Reinforcement learning gyms for experimenting with domain generalization, domain adaptation, and robustness to domain shift