[Bug Report] The reward obtained is not as described in the document

[Seraphli]

Describe the bug Run manual_control.py to play MiniGrid-DoorKey-5x5-v0, and the final reward is 0.97, not 1.0.

Mission: use the key to open the door and then get to the goal
pressed pageup
step=1, reward=0.00
pressed up
step=2, reward=0.00
pressed right
step=3, reward=0.00
pressed  
step=4, reward=0.00
pressed up
step=5, reward=0.00
pressed up
step=6, reward=0.00
pressed right
step=7, reward=0.00
pressed up
step=8, reward=0.00
pressed up
step=9, reward=0.97
terminated!

System Info minigrid==2.1.0

Checklist

• [x] I have checked that there is no similar issue in the repo (required)

[Seraphli]

I check the code and the reward is changed to 1 - 0.9 * (self.step_count / self.max_steps). Maybe the document needs to be updated.

[pseudo-rnd-thoughts]

@Seraphli What documentation where needs updating?

[Seraphli]

For example, https://minigrid.farama.org/environments/minigrid/EmptyEnv/

[pseudo-rnd-thoughts]

@Seraphli Would you be able to update all of the environment's documentation? It looks like all of the documentation is incorrect This just means modifying the docstring in each of the environment classes in the ## rewards section with a correct sentence

[Seraphli]

Yes, I can make a pull request. How about changing it to this? A reward of '1 - 0.9 * (step_count / max_steps)' is given for success, and '0' for failure.

[Seraphli]

I notice other issues while I'm updating the document.

1. There is no document for GoToObjectEnv. I will add one similar to GoToDoorEnv.
2. @maximecb seems only update the reward function in GoToDoorEnv, which is here https://github.com/Farama-Foundation/Minigrid/blob/master/minigrid/envs/gotodoor.py#L145. but leave the GoToObjectEnv unchanged. I will also update the reward function.
3. In here https://github.com/Farama-Foundation/Minigrid/blob/master/minigrid/envs/gotodoor.py#L140-L141, the GoToDoorEnv will terminate when an agent presses toggle. But the document says toggle is unused. I think this is not good for exploration and should be removed. What do you think about it?

[pseudo-rnd-thoughts]

I notice other issues while I'm updating the document.

1. There is no document for GoToObjectEnv. I will add one similar to GoToDoorEnv.
2. @maximecb seems only update the reward function in GoToDoorEnv, which is here https://github.com/Farama-Foundation/Minigrid/blob/master/minigrid/envs/gotodoor.py#L145. but leave the GoToObjectEnv unchanged. I will also update the reward function.
3. In here https://github.com/Farama-Foundation/Minigrid/blob/master/minigrid/envs/gotodoor.py#L140-L141, the GoToDoorEnv will terminate when an agent presses toggle. But the document says toggle is unused. I think this is not good for exploration and should be removed. What do you think about it?

1. That would be great thanks @BolunDai0216 What are your thoughts on 2 and 3? If we do, we just need to bump the version number