Text this: Bowen Chen: SSAT-Adapter: Enhancing Vision-Language Model Few-shot Learning with Auxiliary Tasks