[llvm] [AMDGPU] Fix PromoteAlloca size check of alloca for store (PR #72528)

[Pierre van Houtryve via llvm-commits]

https://github.com/Pierre-vh approved this pull request.

What did the IR look like before the change? I thought it'd just be folded to undef if you'd try to extract/insert past the end of a vector

In any case, change is good, test looks good, just wondering what this fixes exactly

https://github.com/llvm/llvm-project/pull/72528