Issue with rsl_rl train

When running this command :

`python train_rsl_rl.py --env_name=LeapCubeReorient`

I get the follwing error :

```
Experiment name: LeapCubeReorient-20250906-001828
Logs are being stored in: /home/dhanush/fpo_dev/mujoco_playground/learning/logs/LeapCubeReorient-20250906-001828
Checkpoint path: /home/dhanush/fpo_dev/mujoco_playground/learning/logs/LeapCubeReorient-20250906-001828/checkpoints
Device -- cuda:0
Key device -- {CudaDevice(id=0)}
obs_shape: {'privileged_state': (128,), 'state': (57,)}
Asymmetric observation space
JITing reset and step
Done JITing reset and step
Traceback (most recent call last):
File "/home/dhanush/fpo_dev/mujoco_playground/learning/train_rsl_rl.py", line 247, in
app.run(main)
File "/home/dhanush/fpo_dev/mujoco_playground/.venv/lib/python3.11/site-packages/absl/app.py", line 316, in run
_run_main(main, args)
File "/home/dhanush/fpo_dev/mujoco_playground/.venv/lib/python3.11/site-packages/absl/app.py", line 261, in _run_main
sys.exit(main(argv))
^^^^^^^^^^
File "/home/dhanush/fpo_dev/mujoco_playground/learning/train_rsl_rl.py", line 178, in main
runner = OnPolicyRunner(brax_env, train_cfg_dict, logdir, device=device)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/dhanush/fpo_dev/rsl_rl/rsl_rl/runners/on_policy_runner.py", line 44, in __init__
self.cfg["obs_groups"] = resolve_obs_groups(obs, self.cfg["obs_groups"], default_sets)
~~~~~~~~^^^^^^^^^^^^^^
KeyError: 'obs_groups'
```

Note : the rsl_rl fork I am using is an exact clone of the original rsl_rl repo

```
Using Python 3.11.13 environment at: /home/dhanush/fpo_dev/mujoco_playground/.venv
rsl-rl-lib 3.0.1 /home/dhanush/fpo_dev/rsl_rl
playground 0.0.5 /home/dhanush/fpo_dev/mujoco_playground
```

Thanks!