Fix/bernoulli

[belerico]

Summary

This PR possibly fixes #147 by introducing a safe Bernoulli distribution, in which the mode is computed in a TF-like fashion (https://github.com/tensorflow/tensorflow/blob/e9552f222b3237aafa4773e3f96a8cefb3ebaf95/tensorflow/python/ops/distributions/bernoulli.py#L160)

Type of Change

Please select the one relevant option below:

• Bug fix (non-breaking change that solves an issue)

Checklist

Please confirm that the following tasks have been completed:

• [x] I have tested my changes locally and they work as expected. (Please describe the tests you performed.)
• [x] I have added unit tests for my changes, or updated existing tests if necessary.
• [x] I have updated the documentation, if applicable.
• [x] I have installed pre-commit and run locally for my code changes.

Screenshots or Visuals (Optional)

If applicable, please provide screenshots, diagrams, graphs, or videos of the changes, features or the error.

To prove this i've run a Walker Walk experiment with the following command:

MUJOCO_GL=egl python sheeprl.py exp=dreamer_v3_dmc_walker_walk.yaml fabric.accelerator=gpu fabric.precision=bf16-mixed env.num_envs=4 env.max_episode_stesp=-1

on a Pop-OS laptop with Ubuntu 22.04, an Intel i5-13500H and a NVIDIA RTX 4060 8GB

The experiment has run for more than 8h before being manually stopped.

Thank you for your contribution! Once you have filled out this template, please ensure that you have assigned the appropriate reviewers and that all tests have passed.