[llvm] [AMDGPU][GlobalISel] Properly handle lane op lowering for larger vector types (PR #132358)

[Vikram Hegde via llvm-commits]

vikramRH wrote:

> @vikramRH I've reverted this PR because of builtbot failures. I am wondering if some tests need updating because this PR was created a while ago. Anyway, I'm happy to try your revised patch. Thanks!

@kazutakahirata , thanks, I will raise a revised PR

https://github.com/llvm/llvm-project/pull/132358