fixed executor issue using wrong task attribute and implementation call

synchros2/synchros2/executors.py

@@ -670,6 +670,10 @@ def bind(self, callback_group: rclpy.callback_groups.CallbackGroup, thread_pool:
             self._logger.debug(f"Binding {callback_group_name} to thread pool #{thread_pool_index}...")
             self._callback_group_affinity[callback_group] = thread_pool
 
+    def _spin_once_impl(self, *args: typing.Any, **kwargs: typing.Any) -> None:
+        """Spin the executor once (for Jazzy and beyond)."""
+        self._do_spin_once(*args, **kwargs)
+
     def _do_spin_once(self, *args: typing.Any, **kwargs: typing.Any) -> None:
         with self._spin_lock:
             try:
@@ -787,9 +791,15 @@ def shutdown(self, timeout_sec: typing.Optional[float] = None) -> bool:
             with self._spin_lock:
                 # rclpy.executors.Executor base implementation leaves tasks
                 # unawaited upon shutdown. Do the housekeepng.
-                known_tasks = [
-                    AutoScalingMultiThreadedExecutor.Task(task, entity, node) for task, entity, node in self._tasks
-                ] + list(self._work_in_progress)
+                if hasattr(self, "_tasks"):  # ROS 2 Humble
+                    known_tasks = [
+                        AutoScalingMultiThreadedExecutor.Task(task, entity, node) for task, entity, node in self._tasks
+                    ] + list(self._work_in_progress)
+                else:  # ROS 2 Jazzy and beyond
+                    known_tasks = [
+                        AutoScalingMultiThreadedExecutor.Task(task, task_data.source_entity, task_data.source_node)
+                        for task, task_data in self._pending_tasks.items()
+                    ] + list(self._work_in_progress)
                 for task in known_tasks:
                     task.cancel()
         return done