[Bug Report] gymnasium 0.29.1 not compatible

[Fisher-Wang]

If you are submitting a bug report, please fill in the following details and use the type [Bug].

Describe the bug

As title.

Steps to reproduce

conda create -n metasim_jax_tmp_py310 python=3.10 -y
conda activate metasim_jax_tmp_py310
uv pip install ".[mjx]" gymnasium==0.29.1

Run the script below

import metasim
from gymnasium import make_vec
env_id = "RoboVerse/reach_origin"
env = make_vec(
    env_id,
    robots=['franka'],
    simulator='mjx',
    num_envs=4,
    headless=True,
    cameras=[],
    device='cuda',
)

You got the error

File ~/miniforge3/envs/rvrl_rv/lib/python3.11/site-packages/gymnasium/envs/registration.py:994, in make_vec(id, num_envs, vectorization_mode, vector_kwargs, wrappers, **kwargs)
    989     env = gym.experimental.vector.SyncVectorEnv(
    990         env_fns=[_create_env for _ in range(num_envs)],
    991         **vector_kwargs,
    992     )
    993 elif vectorization_mode == "async":
--> 994     env = gym.experimental.vector.AsyncVectorEnv(
    995         env_fns=[_create_env for _ in range(num_envs)],
    996         **vector_kwargs,
    997     )
    998 elif vectorization_mode == "custom":
...
--> 395     chunk = read(handle, remaining)
    396     n = len(chunk)
    397     if n == 0:

Checklist

• I have checked that there is no similar issue in the repo (required)
• I have checked that the issue is not in the simulator backends (IsaacLab, IsaacGym, MuJoCo, etc.) and is related to the RoboVerse repo

Acceptance Criteria

Add the criteria for which this task is considered done. If not known at issue creation time, you can add this once the issue is assigned.

• support gymnasium==0.29.1 (recommended)
• or specify gymnasium==1.2.1 in pyproject.toml

[Fisher-Wang]

using gym.make instead of gym.make_vec doesn't solve the issue. This will only return the single env