SWDEV-497145 - Use rocr copyOnEngine API for staged copies

- Refactor blit code and clean ASAN instrumentation
- Use unified function for rocr copy
- Enable shader copy path for unpinned writeBuffer/readBuffer paths
- Set GPU_FORCE_BLIT_COPY_SIZE=16 which means we will use BLIT copy for
  pinned copies or unpinned H2D/D2H copies < 16KB

Change-Id: I42045cca79234b340dbf53dafb93044199736ae4

rocclr/device/rocm/rockernel.cpp

@@ -57,8 +57,7 @@ bool LightningKernel::postLoad() {
   }
   kernargSegmentAlignment_ =
       amd::alignUp(std::max(kernargSegmentAlignment_, 128u),
-                   device().info().globalMemCacheLineSize_ > 0 ?
-                       device().info().globalMemCacheLineSize_ : 64);
+                   device().info().globalMemCacheLineSize_);
 
   // Set the workgroup information for the kernel
   workGroupInfo_.availableLDSSize_ = device().info().localMemSizePerCU_;