[llvm] [NVPTX] Vectorize and lower 256-bit global loads/stores for sm_100+/ptx88+ (PR #139292)

[Artem Belevich via llvm-commits]

================
@@ -0,0 +1,728 @@
+; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5
----------------
Artem-B wrote:

I'd rename the test to reflect that it's for 256-bit loads and stores. E.g. `load-store-256-bit.ll`

https://github.com/llvm/llvm-project/pull/139292