SWDEV-486602 - Optimize HSA callback performance

- Don't generate callbacks for HIP events
- Don't process profiling info in the callback for HIP events
- Wait for CPU status update of the submitted commands
every 50 calls. That will allow to drain the commands and
destroy HSA signals.

Change-Id: Ib601a350e7e7c2b6c6209a172385389baccf73a9

rocclr/platform/command.cpp

@@ -366,7 +366,9 @@ void Command::enqueue() {
     ScopedLock sl(queue_->vdev()->execution());
     queue_->FormSubmissionBatch(this);
 
-    if (type() == CL_COMMAND_MARKER || type() == 0 || type() == CL_COMMAND_TASK) {
+    // Enqueue flushes, except profiling markers to avoid frequent expensive callbacks
+    if (((type() == 0) && profilingInfo().batch_flush_) ||
+        (type() == CL_COMMAND_MARKER) || (type() == CL_COMMAND_TASK)) {
       // The current HSA signal tracking logic requires profiling enabled for the markers
       EnableProfiling();
       // Update batch head for the current marker. Hence the status of all commands can be