[clang] [llvm] [AMDGPU] Use a general form of intrinsic for tensor load/store (PR #182334)

[Krzysztof Drewniak via llvm-commits]

krzysz00 wrote:

... Actually, you'll probably want to get the MLIR wrapper around the intrinsic as well

https://github.com/llvm/llvm-project/pull/182334