[llvm] [NVPTX] Add tex.grad.cube{array} intrinsics (PR #77693)

[Artem Belevich via llvm-commits]

================
@@ -1,12 +1,12 @@
 # RUN: %python %s --target=cuda --tests=suld,sust,tex,tld4 --gen-list=%t.list > %t-cuda.ll
-# RUN: llc -mcpu=sm_20 %t-cuda.ll -verify-machineinstrs -o - | FileCheck %t-cuda.ll
----------------
Artem-B wrote:

We should still have the CPU set explicitly. `sm_60` would probably be the reasonable oldest non-deprecated arch.

https://github.com/llvm/llvm-project/pull/77693