Opportunistic Routing in Wireless Communications via Learnable State-Augmented Policies