fix(dataset): correct variable for language detection

- Change `bits` to `self._data_bits` for language detection
- This fixes the TypeError when opening PDF files

magic_pdf/data/dataset.py

@@ -249,7 +249,7 @@ class ImageDataset(Dataset):
         elif lang == 'auto':
             from magic_pdf.model.sub_modules.language_detection.utils import \
                 auto_detect_lang
-            self._lang = auto_detect_lang(bits)
+            self._lang = auto_detect_lang(self._data_bits)
             logger.info(f'lang: {lang}, detect_lang: {self._lang}')
         else:
             self._lang = lang