Migrate gym to gymnasium

[montplaisir]

Proposed change(s)

Merge these PRs related to gymnasium:

• Pettingzoo wrapper changed to gymnasium interface #6211
• Upgrade to Gymnasium interface #6203
• Update Gym to Gymnasium #6166
• Update unity_pettingzoo_base_env.py #6136
• Upgraded from gym to Gymnasium #6060

Migrate from Gym to Gymnasium (gymnasium>=0.26.0)
Upgrade Pettingzoo to 1.24.0
Relax the dependencies for Numpy

Co-authored-by: @alexander-zap
Co-authored-by: @vadim0x60
Co-authored-by: @AmineAndam04
Co-authored-by: @Zach-Attach

Types of change(s)

• Bug fix
• New feature
• Code refactor
• Breaking change
• Documentation update
• Other (please describe)

Checklist

• Added tests that prove my fix is effective or that my feature works
• Updated the changelog (if applicable)
• Updated the documentation (if applicable)
• Updated the migration guide (if applicable)

[u-pr]

💡 Harness Review

The API migration is mostly coherent, but a few integration gaps remain: the new PettingZoo reset(seed=...) path does not actually seed anything, and some shipped examples/docs still target contracts this PR removes.

The PettingZoo migration is incomplete across the published examples/docs. com.unity.ml-agents/Documentation~/Python-PettingZoo-API.md and ml-agents-envs/colabs/Colab_PettingZoo.ipynb still unpack env.last() as (observation, reward, done, info), while this PR changes the wrapper to return (observation, reward, terminated, truncated, info). The generated API doc also still advertises the old reset()/seed()/render(mode=...) surface. After this merges, users following the repo’s own guides will hit unpacking errors and try methods/signatures that no longer exist. Please update those docs/examples in the same change.

_{Reviewed commit e4ba566f}

_{🤖 Helpful? 👍/👎}