mir_build: Prefer exhaustive destructuring in MatchPairTree::for_pattern

[Zalathar]

This is a bit more verbose, but clearly indicates which variant fields are being ignored, which is helpful for this sort of exhaustive lowering step.

Inspired by #150681 (comment).

There should be no change to compiler behaviour.

[rustbot]

Some changes occurred in match lowering

cc @Nadrieril

[rustbot]

r? @fee1-dead

rustbot has assigned @fee1-dead.
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

[Nadrieril]

I dunno, I find this kind of thing adds more friction than help... If I add a field to an enum like this it's always only going to get used at one or two use sites. thir::PatKind is used in quite a few places too. I'm the kind of person who actively does the exact opposite change than this PR i.e. adding , .. to save us from future pain.

[Zalathar]

Hmm, you might be right.

This is not a big deal either way, so I'm happy to close.

[Nadrieril]

My preference would be to leave things as-is yea