This PR contains several unrelated zero-impact formatting changes; please remove them.

More generally, why are we following point releases of GHC like this? I can understand doing major version bumps to keep up with the times, and the occasional minor bump to get an important feature or compatibility with an LTS snapshot or something, but I don't want to do this every month if we don't have a reason to.

This PR contains several unrelated zero-impact formatting changes; please remove them.

Sure. I have removed all unintentional formatting changes already (vscode did them automatically without asking me, bad boy).

More generally, why are we following point releases of GHC like this?

Idk about any existing conventions. The previous bump wasn't a major one and it let me to believe that this is the way.

I can understand doing major version bumps to keep up with the times, and the occasional minor bump to get an important feature or compatibility with an LTS snapshot or something, but I don't want to do this every month if we don't have a reason to.

I respect your time and effort, and will follow any guideline you decide to come up with. I'd also appreciate if a decision criteria are not subjective (e.g. an "importance of a certain feature or compatibility" is usually a relative measure).

Please close this PR if you feel like its not worth your time reviewing and answering it.

Thank you!

The previous bump was made as we were previously on a nightly package set and a GHC version that wasn't supported by HLS (if I recall correctly, at least). Like mentioned, unless there's a super important bugfix or if some incompatibility, keeping up with minor releases can become a bit tedious.

Moving forward, I think we can mark this PR as a WIP until we find the need to bump the minor version or it gets superseded by another version bump.

Btw, GHC 9.2.7 is not on the list of supported versions by HLS.

Closing as this doesn't seem like we'll be merging.