Custom exception class for errors associated with the Google GenAI API.
Google Generative AI Embeddings.
While gemini-embedding-2-preview natively supports multimodal inputs
(text, images, video, audio, and PDFs) via the Google GenAI SDK, the
LangChain Embeddings interface (embed_query / embed_documents)
currently only accepts text. For multimodal embedding use cases in the
meantime, use the `Google GenAI SDK directly.
