[llvm] [AMDGPU] Use subtarget feature for flat offset bit width instead of arch checks (PR #183742)

[Mariusz Sikora via llvm-commits]

================
@@ -235,6 +236,8 @@ class AMDGPUSubtarget {
     return AddressableLocalMemorySize;
   }
 
+  unsigned getFlatOffsetBitWidth() const { return FlatOffsetBitWidth; }
----------------
mariusz-sikora-at-amd wrote:

It is not, I will remove it.

https://github.com/llvm/llvm-project/pull/183742