OCR parser update

[tenzin3]

Description

Currently, the OCR parser stores annotation information in the metadata.json file. However, this annotation data should instead be returned alongside the parsed pecha to the OpenPecha backend, where it can be properly stored.

Expected Output

The OpenPecha backend should be capable of storing both the pecha and all relevant annotation information provided by the OCR parser.

Implementation Steps

• Create BDRC annotation models.
• Modify the HOCR parser to return annotation data.
• Remove the logic that stores annotation information in metadata.json.
• Modify the Google Vision parser to return annotation data.
• Ensure annotation data is not stored in metadata.json.

Reviewer

• @ta4tsering

[tenzin3]

There is work id and scan id but this is not consistent. This should be in pecha metadata.