Out-of-the-box Parameter Control for Evolutionary and Swarm-based Algorithms with Distributed Reinforcement Learning