Extend branches_sharing_code to match arms with a shared tail

[sylvestre]

This extends branches_sharing_code to also fire on match expressions, not just if/else.

What it does

When every arm of a match is a block ending in the same trailing expression, and the match is itself in tail position, that expression can be hoisted out below the match:

match mode {
    Mode::A => { a(); Ok(()) }
    Mode::B => { b(); Ok(()) }
}

becomes

match mode {
    Mode::A => { a(); }
    Mode::B => { b(); }
}
Ok(())

This came up from a real cleanup (uutils/coreutils#13129) where a function repeated Ok(()) in every arm.

Guards (to keep it correct and quiet)

• Every arm body must be a block with a trailing expression; all tails must be HirEq.
• The tail must not reference any binding introduced inside the arm (pattern bindings or block lets).
• Trivial () tails and the all-arms-identical case are skipped.
• A machine-applicable fix is only offered when the match is the trailing expression of its enclosing block, so appending the expression after the match is always valid. When the value is used elsewhere (e.g. bound in a let), the lint stays quiet.

Lintcheck

Ran filtered to this lint:

• Default crate set (25 crates): 5 new match hits (all in wasmi), 0 false positives.
• Top 200 most-downloaded crates: 3 new match hits (cc, regex-syntax ×2), 0 false positives.

All inspected hits are genuine and the fixes compile.

Notes

• This does not catch the exact uutils PR above: there one arm's tail was an if/else with Err(..) in one branch, so the shared Ok(()) wasn't a direct arm tail — that needs early-return restructuring, out of scope here. It catches the simpler, common sibling of that pattern.
• branches_sharing_code is nursery, so this is opt-in.

changelog: [branches_sharing_code]: also lint match expressions whose arms end with the same expression

[rustbot]

r? @llogiq

rustbot has assigned @llogiq.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

Why was this reviewer chosen?

The reviewer was selected based on:

• Owners of files modified in this PR: 8 candidates
• 8 candidates expanded to 8 candidates
• Random selection from Jarcho, dswij, llogiq, samueltardieu

[github-actions]

Lintcheck changes for 4dd09d6

Lint	Added	Removed	Changed
clippy::branches_sharing_code	8	0	2

This comment will be updated if you push new changes

[llogiq]

I like the extension. The only thing I'd like to see is a few test cases with comments in the match arms (or even before the =>).

View changes since this review

[rustbot]

This PR was rebased onto a different master commit. Here's a range-diff highlighting what actually changed.

Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers.

[sylvestre]

@llogiq done! thanks

[sylvestre]

@llogiq can we merge it now? :)
Thanks sorry for the pressure