Add comments from paper

Adopt DDPG to multiagent-particle-envs

Make possible to switch env to continuous action space

Improve logging output

Adopt DDPG to "simple" scenario from multiagent-particle-envs

DDPG - tensorflow implementation

Add torch ddpg implementation

simple tag: add done()

Simple: add done()

SSL adjust reward

SSL adjust observation()

Edit reward() in simple-scenario

Turn off verbose logging while rendering

BaseScenario - add empty done()-method

Fix observation