[llvm] [AMDGPU][GlobalISel] Select s16 G_MERGE_VALUES into wider scalars (PR #207999)

[Keshav Vinayak Jha via llvm-commits]

keshavvinayak01 wrote:

> I'm surprised we don't have e2e impacts here, but maybe this just doesn't come up much?

Yeah, I do need to add an e2e test then.

https://github.com/llvm/llvm-project/pull/207999