Enhance training and logging functionality with accelerator support

- Added support for multi-GPU training by introducing an `accelerator` parameter in training functions.
- Updated `update_policy` to handle gradient updates based on the presence of an accelerator.
- Modified logging to prevent duplicate messages in non-main processes.
- Enhanced `set_seed` and `get_safe_torch_device` functions to accommodate accelerator usage.
- Updated `MetricsTracker` to account for the number of processes when calculating metrics.
- Introduced a new feature in `pyproject.toml` for the `accelerate` library dependency.

pyproject.toml

@@ -125,6 +125,7 @@ hilserl = ["lerobot[transformers-dep]", "gym-hil>=0.1.11", "lerobot[grpcio-dep]"
 
 # Features
 async = ["lerobot[grpcio-dep]", "matplotlib>=3.10.3"]
+accelerate = ["accelerate>=1.10.0"]
 
 # Development
 dev = ["pre-commit>=3.7.0", "debugpy>=1.8.1", "lerobot[grpcio-dep]", "grpcio-tools==1.73.1"]