v1.3.0

Added greedy agent as strong baseline (wins with search depth 2), added testing for Tianshou & internal game states, fixed edge cases in game logic & UI/interactive play.
Full changelog: https://github.com/elliottower/gobblet-rl/commits/