[llvm] [X86][GlobalIsel] Add support for G_UMIN/G_UMAX/G_SMIN/G_SMAX (PR #160247)

[Evgenii Kudriashov via llvm-commits]

================
@@ -147,6 +147,8 @@ X86LegalizerInfo::X86LegalizerInfo(const X86Subtarget &STI,
         });
   }
 
+  getActionDefinitionsBuilder({G_UMIN, G_UMAX, G_SMIN, G_SMAX}).lower();
----------------
e-kud wrote:

These are integer types, can we clamp them between `s8` and `s64`  and widen to the next power of 2?

https://github.com/llvm/llvm-project/pull/160247