Using Diffusers for reinforcement learning

Support for one RL model and related pipelines is included in the experimental source of diffusers.

To try some of this in colab, please look at the following example:

• Model-based reinforcement learning on Colab