Error-Aware Imitation Learning from Teleoperation Data for Mobile Manipulation