Intersectional Fairness in Reinforcement Learning with Large State and Constraint Spaces