fix: Use atomic for cross-thread access in thread_pool and test

TSan detected data races:
- thread_pool::m_IsShuttingDown was accessed without synchronization
- coroutine_task_test value variable was modified on thread pool and
  read on main thread without synchronization

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

Author: mhaynie

cpp/include/mh/concurrency/thread_pool.inl

@@ -8,13 +8,15 @@
 
 #include <utility>
 
+#include <atomic>
+
 namespace mh
 {
 	namespace detail::thread_pool_hpp
 	{
 		struct thread_data
 		{
-			bool m_IsShuttingDown = false;
+			std::atomic<bool> m_IsShuttingDown = false;
 
 			mh::dispatcher m_Dispatcher{ false };
 			std::vector<std::thread> m_Threads;

test/coroutine_task_test.cpp

@@ -150,9 +150,9 @@ TEST_CASE("task - exceptions in discarded tasks")
 {
 	mh::thread_pool tp(2);
 
-	int value = 0;
+	std::atomic<int> value = 0;
 	// Intentionally discarding the task - cast to void to suppress nodiscard warning
-	(void)[](mh::thread_pool& tp, int& val) -> mh::task<>
+	(void)[](mh::thread_pool& tp, std::atomic<int>& val) -> mh::task<>
 	{
 		co_await tp.co_add_task();
 		co_await tp.co_delay_for(2s);