[llvm] [InstCombine] Handle trunc i1 pattern in eq-of-parts fold (PR #112704)

[via llvm-commits]

goldsteinn wrote:

LGTM no regressions on Yingwei's benchmarks.

Also note, some further missing patterns: https://godbolt.org/z/b79Gnrxnx

https://github.com/llvm/llvm-project/pull/112704