Training and Finetuning Multimodal Embedding & Reranker Models with Sentence Transformers