I\'ve found out that using the TEXT_DETECTION mode provides better OCR results than the DOCUMENT_TEXT_DETECTION mode. Is there a way to receive the confidence level also on
