[llvm] [AMDGPU] Promote allocas used by phi/select (PR #221870)

[Matt Arsenault via llvm-commits]

================
@@ -0,0 +1,236 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 6
+; RUN: opt -S -mtriple=amdgpu-unknown-amdhsa -passes=amdgpu-promote-alloca-to-vector < %s | FileCheck %s
----------------
arsenm wrote:

```suggestion
; RUN: opt -S -mtriple=amdgpu7.00-unknown-amdhsa -passes=amdgpu-promote-alloca-to-vector < %s | FileCheck %s
```

https://github.com/llvm/llvm-project/pull/221870