Never request reviews on staging-esque branches

[sternenseemann]

PRs from staging-next usually never cause ofborg to request reviews from package maintainers because the change count will be high enough even in the beginning to prevent it. However the key insight is that we don't want to not request reviews because there would be too many, but rather the changes have already been reviewed (before being merged into staging for example).

There are in fact branches similar to staging-next, like haskell-updates, where ofborg will still request reviews, but unnecessarily so, because the branch mostly contains already reviewed and merged PRs. In https://github.com/NixOS/nixpkgs/pull/139967 in particular you can see a good example of this, including Damien Cassou fighting the ofborg request for review…

This commit adds a new condition to the review requesting logic and checks if the head of the PR is a staging branch (i. e. matches staging-next* or haskell-updates and is in the NixOS/nixpkgs repository). This should prevent such cases reliably in the future. The list of branches can be extended easily.

---

Help with testing / reviewing would be appreciated, I can't really run ofborg locally to test this change…

[Profpatsch]

Bump