[llvm] [ExpandMemCmp] Optimize ExpandMemCmp to reduce instruction count on x86 (PR #69609)

[Igor Kirillov via llvm-commits]

igogo-x86 wrote:

> Wouldn't that byte swap the zeros that were added by the zext into the LSBs? I don't know this code, but that seems weird.

Yeah, they would, but since they are zeroes, they wouldn't affect the comparison result.

https://github.com/llvm/llvm-project/pull/69609