SWDEV-508279 - Improve HIP event profiling

There are 2 functional changes to this patch:
* Use GPU timing for internal markers for HIP.
* Measure CPU time closer to GPU timer, to reduce delta between GPU/CPU timestamp measurements.

There are some smaller non-functional updates:
* waifForFence -> waitForFence typo
* Remove unused drmProfiling

Change-Id: I4c5fa600a842ab60e454888779edcac8449a902a


[ROCm/clr commit: 179801a750]

projects/clr/hipamd/src/hip_event.cpp

@@ -224,7 +224,9 @@ hipError_t Event::recordCommand(amd::Command*& command, amd::HostQueue* stream,
       releaseFlags = amd::Device::kCacheStateInvalid;
     }
     // Always submit a EventMarker.
-    command = new hip::EventMarker(*stream, !kMarkerDisableFlush, true, releaseFlags, batch_flush);
+    constexpr bool kMarkerTs = true;
+    command =
+        new hip::EventMarker(*stream, !kMarkerDisableFlush, kMarkerTs, releaseFlags, batch_flush);
   }
   return hipSuccess;
 }