[llvm] [AMDGPU] Do not put memory instructions in *ALU SchedGroups (PR #162560)

[Matt Arsenault via llvm-commits]

================
@@ -2394,18 +2394,18 @@ bool SchedGroup::canAddMI(const MachineInstr &MI) const {
   else if (((SGMask & SchedGroupMask::ALU) != SchedGroupMask::NONE) &&
            (TII->isVALU(MI) || TII->isMFMAorWMMA(MI) || TII->isSALU(MI) ||
             TII->isTRANS(MI)))
-    Result = !(MI.mayLoad() || MI.mayLoad());
+    Result = !(MI.mayLoad() || MI.mayStore());
----------------
arsenm wrote:

```suggestion
    Result = !MI.mayLoadOrStore();
```

https://github.com/llvm/llvm-project/pull/162560