[llvm] [AMDGPU] Add scaffolding for ML focused scheduling strategy (PR #169616)

[Matt Arsenault via llvm-commits]

================
@@ -636,6 +637,11 @@ static ScheduleDAGInstrs *createSIMachineScheduler(MachineSchedContext *C) {
   return new SIScheduleDAGMI(C);
 }
 
+static bool isMLWorkload(const Function &F) {
+  Attribute WorkloadAttr = F.getFnAttribute("amdgpu-workload-type");
----------------
arsenm wrote:

Just use the existing scheduler override attribute? 

https://github.com/llvm/llvm-project/pull/169616