Add reward/loss over time graphs for actor and critic networks

[generic-github-user]

The actor reward graph should display both the predicted loss generated by the critic network (equivalent to the actor optimization loss) and the actual loss once the training episode is complete.

[generic-github-user]

Also display the current reward at each point in the game (although only the reward value at the end of the game is actually used in training).

[quantuminformation]

Would be awesome, like deepminds SC2 demo.