Markov Decision Process in Rust (WIP)

One great lecture to follow. Also Tram problem is taken from the lecture:

Link to lecture

Goals:

1. Simple Markov Decision Process
2. Wait and see if the Rust ecosystem for NN is mature. See how to incorporate more advanced reinforcement learning stuff.
3. Simplify the process.
4. Better documentation.