Update on "[Executorch][llama] Set # of threads to use performant cores"

When using all cores, slower ones are dragging the performance down by blocking
large cores. Perhaps when we have uarch specific implementation, we may not
need this, but this tool is useful in general until we have better API

Differential Revision: [D54766071](https://our.internmc.facebook.com/intern/diff/D54766071/)

[ghstack-poisoned]

Author: kimishpatel

backends/xnnpack/threadpool/threadpool.cpp

@@ -45,8 +45,8 @@ size_t ThreadPool::get_thread_count() const {
 }
 
 bool ThreadPool::_unsafe_reset_threadpool(uint32_t new_thread_count) {
-  // No need to do anything if the count is same
-  if (new_thread_count == get_thread_count()) {
+  // No need to do anything if the count is same or 0
+  if (new_thread_count == get_thread_count() || new_thread_count == 0) {
     return true;
   }