[Bug Report] AntMaze envs ignore seed parameters

[DT6A]

Hello! Seems like there are some problems with AntMaze tasks in terms of setting seeds. The following code:

import numpy as np
import random
import gym
import d4rl

np.random.seed(0); random.seed(0)
env = gym.make("antmaze-umaze-v1", seed=0)
env.seed(0); env.action_space.seed(0); env.observation_space.seed(0)
env.reset()

np.random.seed(0); random.seed(0)
env = gym.make("antmaze-umaze-v1", seed=0)
env.seed(0); env.action_space.seed(0); env.observation_space.seed(0)
env.reset()

outputs different initial states which is not good.

There was an issue https://github.com/Farama-Foundation/D4RL/issues/23 about the same problem for HalfCheetah tasks. Did the fix influenced AntMaze?

[Penguin0007]

Same here.