Text this: VLFSE: Enhancing visual tracking through visual language fusion and state update evaluator