fixed recurrent prev_action issue

ml-agents/mlagents/trainers/agent_processor.py

 
             action_mask = stored_decision_step.action_mask
             prev_action = self.policy.retrieve_previous_action([global_id])
+            for _prev_act_type, _prev_act in prev_action.items():
+                prev_action[_prev_act_type] = _prev_act[0, :]
             experience = AgentExperience(
                 obs=obs,
                 reward=step.reward,

ml-agents/mlagents/trainers/optimizer/tf_optimizer.py

             ]
             feed_dict[self.memory_in] = [np.zeros((self.m_size), dtype=np.float32)]
         if self.policy.prev_action is not None:
-            feed_dict[self.policy.prev_action] = batch["prev_action"]
+            feed_dict[self.policy.prev_action] = batch["prev_discrete_action"]
 
         if self.policy.use_recurrent:
             value_estimates, policy_mem, value_mem = self.sess.run(

ml-agents/mlagents/trainers/sac/optimizer_tf.py

         else:
             feed_dict[policy.output] = batch["discrete_action"]
             if self.policy.use_recurrent:
-                feed_dict[policy.prev_action] = batch["prev_action"]
+                feed_dict[policy.prev_action] = batch["prev_discrete_action"]
             feed_dict[policy.action_masks] = batch["action_mask"]
         if self.policy.use_vec_obs:
             feed_dict[policy.vector_in] = batch["vector_obs"]

ml-agents/mlagents/trainers/tests/tensorflow/test_simple_rl.py

     _check_environment_trains(env, {BRAIN_NAME: config}, success_threshold=0.5)
 
 
-@pytest.mark.parametrize("use_discrete", [True])
+@pytest.mark.parametrize("use_discrete", [True, False])
 def test_recurrent_ppo(use_discrete):
     env = MemoryEnvironment([BRAIN_NAME], use_discrete=use_discrete)
     new_network_settings = attr.evolve(