Fix dangling reference in async lambda captures in thread_helper.hpp (#104)

* Initial plan

* Fix dangling reference in thread_helper.hpp async lambda captures

Co-authored-by: xipingyan <47800531+xipingyan@users.noreply.github.com>

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: xipingyan <47800531+xipingyan@users.noreply.github.com>

Author: Copilot

src/cpp/src/module_genai/utils/thread_helper.hpp

@@ -17,24 +17,24 @@ namespace ov::genai::module::thread_utils {
 #endif
 
 #ifdef ENABLE_DYNAMIC_LOAD_MODEL_WEIGHTS
-inline std::future<bool> load_model_weights_async(ov::CompiledModel& compiled_model) {
-    auto load_fun = [&]() -> bool {
+inline std::future<bool> load_model_weights_async(ov::CompiledModel compiled_model) {
+    auto load_fun = [compiled_model]() mutable -> bool {
         PROFILE(pm, "load_model_weights async");
         compiled_model.load_model_weights();
         // infer_request = compiled_model.create_infer_request();
         return true;
     };
-    return std::async(std::launch::async, load_fun);
+    return std::async(std::launch::async, std::move(load_fun));
 }
 
-inline std::future<bool> release_model_weights_async(ov::CompiledModel& compiled_model, ov::InferRequest infer_request) {
-    auto release_fun = [&]() -> bool {
+inline std::future<bool> release_model_weights_async(ov::CompiledModel compiled_model, ov::InferRequest infer_request) {
+    auto release_fun = [compiled_model, infer_request]() mutable -> bool {
         PROFILE(pm, "release_model_weights async");
         infer_request = ov::InferRequest();  // reset infer request to release the reference to the model weights
         compiled_model.release_model_weights();
         return true;
     };
-    return std::async(std::launch::async, release_fun);
+    return std::async(std::launch::async, std::move(release_fun));
 }
 #endif