ENTRAP-VL: A Taxonomic Probe for Dual Contextual Entrainment in Vision-Language Models | Digital Library | PAMCET | PAMCET