Model-Based Transfer Learning for Contextual Reinforcement Learning