Targeted Attacks and Defenses for Distributed Federated Learning in Vehicular Networks