Skip group info mismatch error for broken groups

[pcapriotti]

Commits with a broken group info are now let through if the group was already broken.

https://wearezeta.atlassian.net/browse/WPB-21363

Checklist

• Add a new entry in an appropriate subdirectory of changelog.d
• Read and follow the PR guidelines

[Copilot]

Pull request overview

This PR implements logic to skip group info mismatch errors for MLS conversations that were already in a broken state before the group info diagnostics feature was enabled. The implementation allows systems to gracefully handle pre-existing broken groups without blocking future commits.

Key Changes

• Refactored group state validation to detect and tolerate pre-existing mismatches
• Added logic to check if a stored group already has a state mismatch before rejecting new commits
• Added comprehensive test coverage for the new behavior

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
services/galley/src/Galley/API/MLS/Message.hs	Updated checkGroupState calls to pass ConvOrSubConv instead of Maybe TeamId
services/galley/src/Galley/API/MLS/GroupInfoCheck.hs	Refactored validation logic to extract pure groupStateMismatch function and added existingGroupStateMismatch to check for pre-existing issues
libs/wire-subsystems/src/Wire/ConversationStore.hs	Added helper function getConvOrSubGroupInfo to retrieve group info for conversations or subconversations
integration/test/Test/MLS.hs	Added test case testGroupInfoAlreadyBroken to verify commits are allowed when group was already broken
changelog.d/2-features/mls-skip-error-for-broken-groups	Added changelog entry documenting the feature

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[battermann]

Is the purpose of the 'do' just for visual structuring?

[pcapriotti]

Usually I add a nested block when there are some bound variables that I don't want to leak into the parent scope, but here in fact it doesn't seem necessary. I don't mind it in any case, but I can remove it if you want.

[battermann]

Yes, makes sense.
Personally, instead of a comment and a do block I would extract it as a function named enableFeatures. But you don't have to change it, it's fine. This is a super nitpick :)