[llvm] [NVPTX] Implement isTruncateFree and isZExtFree for i32/i64 Optimizations (PR #114683)

[Artem Belevich via llvm-commits]

================
@@ -0,0 +1,17 @@
+; RUN: llc -march=nvptx64 < %s | FileCheck %s
+
----------------
Artem-B wrote:

+1 to that. The test should have some code which has to choose between free ops vs an alternative which would be used otherwise (e.g. `i32 trunc(add(i64,i64))` vs `i32 add(trunc(i64), trunc(i64))`)

https://github.com/llvm/llvm-project/pull/114683