[llvm] [VectorCombine] Add Cmp and Select for shuffleToIdentity (PR #92794)

[Simon Pilgrim via llvm-commits]

RKSimon wrote:

Already committed to trunk. I'll probably try to extend it further as 256-bit integer ops are tricky on AVX1, so please report any other perf issues you see.

https://github.com/llvm/llvm-project/pull/92794