Fix set_action_for_agent (#4691)

* fixes

* Update CHANGELOG.md

Co-authored-by: Chris Elion <chris.elion@unity3d.com>

com.unity.ml-agents/CHANGELOG.md

 ### Minor Changes
 #### com.unity.ml-agents / com.unity.ml-agents.extensions (C#)
 #### ml-agents / ml-agents-envs / gym-unity (Python)
+- `ActionSpec.validate_action()` now enforces that `UnityEnvironment.set_action_for_agent()` receives a 1D `np.array`.
 
 ### Bug Fixes
 #### com.unity.ml-agents (C#)

ml-agents-envs/mlagents_envs/base_env.py

         return action
 
     def _validate_action(
-        self, actions: np.ndarray, n_agents: int, name: str
+        self, actions: np.ndarray, n_agents: Optional[int], name: str
     ) -> np.ndarray:
         """
         Validates that action has the correct action dim
             _size = self.continuous_size
         else:
             _size = self.discrete_size
-        _expected_shape = (n_agents, _size)
+        _expected_shape = (n_agents, _size) if n_agents else (_size,)
         if actions.shape != _expected_shape:
             raise UnityActionException(
                 f"The behavior {name} needs an input of dimension "

ml-agents-envs/mlagents_envs/environment.py

             return
         action_spec = self._env_specs[behavior_name].action_spec
         num_agents = len(self._env_state[behavior_name][0])
-        action = action_spec._validate_action(action, num_agents, behavior_name)
+        action = action_spec._validate_action(action, None, behavior_name)
         if behavior_name not in self._env_actions:
             self._env_actions[behavior_name] = action_spec.empty_action(num_agents)
         try: