[llvm] [AMDGPU] Enable vectorization of i8 values. (PR #134934)

[Gheorghe-Teodor Bercea via llvm-commits]

doru1004 wrote:

I am in the process of precommitting the tests: https://github.com/llvm/llvm-project/pull/138801 so if someone can approve that PR I can merge and then rebase this patch so changes are more obvious.

https://github.com/llvm/llvm-project/pull/134934