[WIP] Add stability score calculation with p95 latency and jitter

[Copilot]

Thanks for assigning this issue to me. I'm starting to work on it and will keep this PR's description up to date as I form a plan and make progress.

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Stability Score (variance + p95)</issue_title>
<issue_description>What to add
• p95 latency (or p90): “95% of pings are faster than X ms”
• Jitter / variance: how much the latency fluctuates (stable vs spiky)
• Spike counter: number of pings above a threshold (e.g. > 3s)
• Failure streak / last fail: consecutive failures and recency
• Optional: p95 over the last N minutes (sliding window) vs since start

Why it matters (simple example)
• Model A: Avg 250ms, but p95 6000ms (tons of spikes)
• Model B: Avg 400ms, p95 650ms (boringly consistent)
In real usage, Model B feels faster because it doesn’t randomly stall your agent/IDE.

How it should affect ranking

Add a Stability Score that rewards:
• low p95
• low jitter
• low spike rate
• low failure streak

</issue_description>

<agent_instructions>Please do that in a branch named "stability_score". If you have better ideas for more data and then translate that into a human readable score, or state (with words or emojis) it would be nice</agent_instructions>

Comments on the Issue (you are @copilot in this section)

• Fixes vava-nessa/vavanessa.dev#1

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.