Early stopping

Also written: early_stopping_patience

Ending training when the chosen metric hasn't improved for a set number of evaluations, then reloading the best epoch. It needs `load_best_model_at_end=True`.