build(deps): bump github.com/nspcc-dev/neo-go from 0.78.3 to 0.99.7

[dependabot]

Bumps github.com/nspcc-dev/neo-go from 0.78.3 to 0.99.7.

Release notes

Sourced from github.com/nspcc-dev/neo-go's releases.

Hyalinization

Bugs, terrestrial arthropods with at least six legs. Sometimes we don't notice them, sometimes we ignore them, but they do exist and there are times when they need to be dealt with in one way or another. Most of the bugs fixed in this release have a little less legs than proper insects, they're far from being as critical as the ones fixed in the previous release and yet we consider them annoying enough to warrant this new version of NeoGo to be published. We don't want to see our users fighting these pesky creatures in the wild, better update and have a troublefree experience.

If you're not affected by #2815 (requests for verbose transaction data and application logs failed for some specific mainnet transaction(s)) you may leave the DB as is with this upgrade, if you want it to be solved then please resynchronize.

Behaviour changes:

• "loadtx" VM CLI command now uses transaction system fee value as the default GAS limit for the context, unless --gas is used (previously it wasn't limited, #2816)
• SendRawTransaction RPC client method will now always return transaction hash, even if error occured, this also affects Actor APIs and allows to handle some specific errors (like transaction already present in the mempool) in more appropriate way (#2817)
• Actor and NotaryActor Wait() wrapper methods now handle "already exists" and "already on chain" errors as non-errors, waiting (if needed) and returning proper results (#2817)

Improvements:

• reworked network discoverer/connection manager that has a better picture of the network (especially if it's a small one), doesn't try to make additional connections (even when having less than MinPeers of them) when there are no more real addresses to connect to, establishes stable connections faster in containerized environments and does more connection attempts when MinPeers setting is not satisfied, but there are potential candidates (#2811)
• VM invocation stack handling microoptimization (#2812)
• load* VM CLI commands now accept --gas parameter to set the amount of GAS to be used in this context (#2816)
• better logging during state reset process (#2813)
• subscription-based transaction waiter now handles already accepted transactions faster (~immediately vs waiting for block timeout previously, #2817)
• WSClient can return more specific error on websocket handshake failure if it's provided by server (#2821)
• new MaxWebSocketClients configuration option for the RPC server to control the number of allowed websocket clients, missing or 0 value make the server use the default of 64 which is compatible with the previous behavior (you can use old configurations if you don't care about this option, #2821)

Bugs fixed:

• occasional "loadgo" VM CLI test failures because of timeout (#2810)
• websocket waiter test instability (#2809)
• websocket event unsubscriptions could lead to node deadlock in rare circumstances (#2809)
• exception handler could use improper evaluation stack in some cases (#2812)
• pointer stack items were not handled correctly in "protected" serialization mode leading to inability to deserialize these items back (#2816)
• state reset functionality not working for big chains like mainnet (#2813)

Funambulation

An emergency release fixing mainnet incompatibility at block 2504320 leading to inability to process block 2504813. A couple of other less severe bugs got fixed as well. Unfortunately, this release requires complete resynchronization for mainnet, other networks can use old DBs.

Improvements:

• only transactions requested by the consensus process are forwarded to it now slightly improving performance in some networked scenarios (#2788)
• a complete set of simple array unwrappers in the unwrap library (#2792)
• RPC binding generator can handle simple arrays now with configuration file provided (#2792)
• more user-friendly error messages in some cases (#2797, #2802, #2804)

Bugs fixed:

• DB was not properly closed after state reset (#2791)
• VM stack handling problem leading to lost items after return from call to another contract (#2800)
• "istack" command panic in the VM CLI (#2800)
• failure to decode incorrect contract from ContractManagement state leading to failure to start the node, it'll be ignored now similar to how C# node treats this (#2802)
• subscription to execution events failed if no "state" filter was used along with "container" (#2804)
• Actor using WSClient could deadlock in the transaction waiter during unsubscription (#2804)

Underestimation

It wasn't long since the previous release until a juicy set of features and optimisations came on board. So we've decided to change our release plan and supply one more 3.4.0 compatible version. It comes up with a massive suite of network P2P and broadcast improvements. We've carefully investigated message sending timeouts on benchmarks, reworked the broadcast logic and significantly reduced delays in message sending and the amount of useless traffic between our nodes. For smart-contract developers and RPC client users several new features were added. These are automatic RPC bindings generator, poll-based and websocket-based transactions awaiting and an extended websocket client subscriptions interface. For better debugging experience we've improved state related functionality of VM CLI and added an ability to reset chain state to a particular height.

An important dBFT initialization bug that caused timestamp-related errors on block addition was fixed. Also, those who use NeoGo node with EnableCORSWorkaround setting may want to update the node to be able to handle websocket requests with Origin set in the header. The users of websocket RPC client subscription mechanism are welcomed to move from deprecated subscriptions API to the extended one. Finally, we are kindly asking our users to have a look at our roadmap file where deprecated code removal schedule is attached.

This version is fully compatible with C# node 3.4.0 and does not require resynchronization on upgrade. This time for sure, the next release is planned to be compatible with 3.5.0.

... (truncated)

Changelog

Sourced from github.com/nspcc-dev/neo-go's changelog.

0.99.7 "Hyalinization" (23 Nov 2022)

Bugs, terrestrial arthropods with at least six legs. Sometimes we don't notice them, sometimes we ignore them, but they do exist and there are times when they need to be dealt with in one way or another. Most of the bugs fixed in this release have a little less legs than proper insects, they're far from being as critical as the ones fixed in the previous release and yet we consider them annoying enough to warrant this new version of NeoGo to be published. We don't want to see our users fighting these pesky creatures in the wild, better update and have a troublefree experience.

If you're not affected by #2815 (requests for verbose transaction data and application logs failed for some specific mainnet transaction(s)) you may leave the DB as is with this upgrade, if you want it to be solved then please resynchronize.

Behaviour changes:

• "loadtx" VM CLI command now uses transaction system fee value as the default GAS limit for the context, unless --gas is used (previously it wasn't limited, #2816)
• SendRawTransaction RPC client method will now always return transaction hash, even if error occured, this also affects Actor APIs and allows to handle some specific errors (like transaction already present in the mempool) in more appropriate way (#2817)
• Actor and NotaryActor Wait() wrapper methods now handle "already exists" and "already on chain" errors as non-errors, waiting (if needed) and returning proper results (#2817)

Improvements:

• reworked network discoverer/connection manager that has a better picture of the network (especially if it's a small one), doesn't try to make additional connections (even when having less than MinPeers of them) when there are no more real addresses to connect to, establishes stable connections faster in containerized environments and does more connection attempts when MinPeers setting is not satisfied, but there are potential candidates (#2811)
• VM invocation stack handling microoptimization (#2812)
• load* VM CLI commands now accept --gas parameter to set the amount of GAS to be used in this context (#2816)
• better logging during state reset process (#2813)
• subscription-based transaction waiter now handles already accepted transactions faster (~immediately vs waiting for block timeout previously, #2817)
• WSClient can return more specific error on websocket handshake failure if it's provided by server (#2821)
• new MaxWebSocketClients configuration option for the RPC server to control the number of allowed websocket clients, missing or 0 value make the server use the default of 64 which is compatible with the previous behavior (you can use old configurations if you don't care about this option, #2821)

Bugs fixed:

... (truncated)

Commits

• 92dfec6 Merge pull request #2822 from nspcc-dev/0997
• f10311f CHANGELOG: release 0.99.7
• 3ac76c3 Merge pull request #2821 from nspcc-dev/improve-wsclient-handshake-error
• 327dfb8 rpcsrv: add wsclient error message test for #2818
• 2591c39 rpcsrv: make websocket client limit configurable
• fb09670 rpcclient: extract more detailed server-side on WS connection problem
• ab0b236 Merge pull request #2817 from nspcc-dev/always-return-hash-vub-from-sender
• fd04b2b actor: don't abort waiter on "already exists" error
• 66ddecc Merge pull request #2813 from nspcc-dev/fix-state-reset
• f3ef289 core: check headers at the proper state on state reset
• Additional commits viewable in compare view

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot merge will merge this PR after your CI passes on it
• @dependabot squash and merge will squash and merge this PR after your CI passes on it
• @dependabot cancel merge will cancel a previously requested merge and block automerging
• @dependabot reopen will reopen this PR if it is closed
• @dependabot close will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
• @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)

[dependabot]

Superseded by #248.