VimTS: A Unified Video and Image Text Spotter for Enhancing the Cross-domain Generalization