NoDeath wrapper

[sparisi]

New wrapper that prevents death in states like obstacles or lava, and gives an optional negative reward instead.

Description

The wrapper makes it possible to investigate exploration and safety.
Making it possible to walk on lava (at a cost) would make exploration more interesting because the agent can find some "aggressive" exploration strategies, like walking over lava to explore as much as possible. This would be a scenario that should be avoided in safe RL.

More discussion here
#372

Type of change

Please delete options that are not relevant.

• New feature (non-breaking change which adds functionality)
• This change requires a documentation update

Checklist:

• I have run the pre-commit checks with pre-commit run --all-files (see CONTRIBUTING.md instructions to set it up)
• I have commented my code, particularly in hard-to-understand areas
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes

[sparisi]

I am not sure why the pre-commit fails. If I run it locally, it failed the first time but succeeded after that.

[pseudo-rnd-thoughts]

Could you make these two documentation changes otherwise, looks good to merge to me
@BolunDai0216 Any other comments?

[pseudo-rnd-thoughts]

We run doctest in the CI, you need to update (-1, False) to (-1.0, False)
Run pytest --doctest gymnasium/ to check there aren't any issues

[sparisi]

We run doctest in the CI, you need to update (-1, False) to (-1.0, False) Run pytest --doctest gymnasium/ to check there aren't any issues

Sorry about that, I am not very familiar with pytest and couldn't find where it was failing. Btw, the command was
pytest --doctest-modules minigrid/

Everything seems fine now.