[clang] [flang] [llvm] [mlir] [NVPTX] Validate user-specified PTX version against SM version (PR #174834)

[Guray Ozen via cfe-commits]

================
@@ -146,7 +146,7 @@ def GpuNVVMAttachTarget: Pass<"nvvm-attach-target", ""> {
            /*default=*/"\"sm_50\"",
            "Target chip.">,
     Option<"features", "features", "std::string",
-           /*default=*/"\"+ptx60\"",
+           /*default=*/"\"\"",
----------------
grypp wrote:

nit: you can add a comment here:

```
 Option<"features", "features", "std::string",
           /*default=*/"\"\"",
           "Target features, default is to use the minimum for the SM target.">,
```

https://github.com/llvm/llvm-project/pull/174834