Function understand_image 

pub fn understand_image(
    cfg: &AiConfig,
    image: &[u8],
    mime_type: &str,
    context: ImageContext<'_>,
) -> Result<ImageUnderstanding, Error>

Judges image as a document scan or a general image, and extracts or describes it accordingly, in a single call.

mime_type is the image’s own MIME type (e.g. "image/png") — passed through unmodified into the vision message’s data: URI.