Text this: Distributed Deep Reinforcement Learning Via Split Computing For Connected Autonomous Vehicles