[Feature request] Support for incremental DMM pulls

[giorgi1324]

Zilean scraper is pretty wasteful when it comes to syncing the latest DMM hashlist. From a glance it's downloading the whole hashlist just to see what the latest appends are. It manages to pull on the order of magnitude of 10GB a day.

Looking at the commits on https://github.com/debridmediamanager/hashlists they are append-only. It should be possible for Zilean to keep a local clone of that repo and only do git pulls on it instead of fetching the whole zipped archive on every sync. This will be much faster and and bandwidth-efficient compared to the existing method.

Can this be implemented? Thanks!

[AyushSehrawat]

Did this in my fork AyushSehrawat/zilean. Works well for me. Hopefully prom can merge these change later too 💪🏻

Also added GITHUB_TOKEN for the same usecase.