[llvm] [SLP] Prefer copyable vectorization over alternate opcodes (PR #153684)

[Simon Pilgrim via llvm-commits]

RKSimon wrote:

> MultiSource/Benchmarks/Prolangs-C/simulator - small regressions, caused by early optimizations, should improved with future patches

What do these regressions look like and can they be yak shaved first?

https://github.com/llvm/llvm-project/pull/153684