Set a new embedding model on a fitted BERTopic instance. This enables
transform() after loading when the embedding model was not saved.
bertopic_set_embedding_model(model, embedding_model)The input model (invisibly).
A "bertopic_r" model.
Either a character identifier (e.g., "all-MiniLM-L6-v2") or a Python embedding model object (e.g., a SentenceTransformer instance).