refactor(walker2d_v4): add forward/health reward and ctrl cost to ste…

…p info

This commit adds the 'reward_forward', 'reward_ctrl' and 'reward_survive' to the
info dict that is returned by the `step` method. This was done to
improve consistency with the other environments.

gymnasium/envs/mujoco/walker2d_v4.py

@@ -270,6 +270,9 @@ def step(self, action):
         reward = rewards - costs
         terminated = self.terminated
         info = {
+            "reward_forward": forward_reward,
+            "reward_ctrl": -ctrl_cost,
+            "reward_survive": healthy_reward,
             "x_position": x_position_after,
             "x_velocity": x_velocity,
         }