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/platform/program.cpp

@@ -93,6 +93,9 @@ Program::~Program() {
 void Program::unload() {
   for (const auto& it : devicePrograms_) {
     device::Program& devProgram = *(it.second);
+    if (!devProgram.isCodeObjectLoaded()) {
+      continue;
+    }
     if (!devProgram.runFiniKernels()) {
       LogError("Error running fini kernels for devprogram");
     }