Add support for object and content detection in images

It'd be super helpful if Paste would search inside the contents of PDFs, images, text and other blobs — using Apple Vision APIs. https://developer.apple.com/documentation/vision/classifying_images_for_categorization_and_search