SWDEV-520384 - Improve Fat Binary loading latency (#390)

Init and fini kernel needs to be launched when we load and unload code object. Avoid looping through all kernels within a code object just to run the init and fini kernels. Compiler currently only generates 1 init and fini kernel.

[ROCm/clr commit: cd46294b31]

projects/clr/rocclr/device/rocm/rocprogram.cpp

@@ -278,7 +278,7 @@ bool LightningProgram::createKernels(void* binary, size_t binSize, bool useUnifo
       aKernel->setUniformWorkGroupSize(useUniformWorkGroupSize);
     }
     aKernel->setInternalKernelFlag(internalKernel);
-    kernels()[kernelName] = aKernel;
+    addKernel(aKernel);
   }
   return true;
 }