SWDEV-513197 - Improve launch perf for Device Heap kernels

- If any kernel uses device heap, the launch needs to be preceeded by an
  init kernel, Save on the extra barrier packet launch/flush between the
init heap kernel and user kernel

Change-Id: I8ebc6246188200e5f673dc464bc76a53bcb8b7c6

rocclr/device/rocm/rocblit.cpp

@@ -2537,9 +2537,10 @@ bool KernelBlitManager::initHeap(device::Memory* heap_to_initialize, device::Mem
   setArgument(kernels_[blitType], 2, sizeof(uint), &heap_size);
   setArgument(kernels_[blitType], 3, sizeof(uint), &number_of_initial_blocks);
   address parameters = captureArguments(kernels_[blitType]);
-  result = gpu().submitKernelInternal(ndrange, *kernels_[blitType], parameters, nullptr);
+  result = gpu().submitKernelInternal(ndrange, *kernels_[blitType], parameters,
+                                      nullptr, 0, nullptr, nullptr, true);
   releaseArguments(parameters);
-  gpu().releaseGpuMemoryFence();
+  gpu().Barriers().WaitCurrent();
   return result;
 }