Add RL module

[NoB0]

This includes:

• [x] #212
• [x] #219
• [x] #217
• [x] #231
• [x] #215
• [x] #224
• [ ] #235

The module will provide implementation for actor-critic and DQN reinforcement learning.