Add OCR to datasources that are images containing text

[p6l-richard]

⚠️ Please check that this feature request hasn't been suggested before.

• [X] I searched previous Ideas in Discussions didn't find any similar feature requests.
• [X] I searched previous Issues didn't find any similar feature requests.

🔖 Feature description

• If a user uploads a datasource that's actually an image containing text (think: a pdf file that's a scanned document or an image of a book page)
• superagent OCRs the file to extract the text
• so that the agent can read the extracted text of the datasource

📝 Additional Context

This currently results in the agent saying it doesn't have access to the data source, here's a screen recording: https://discord.com/channels/1110910277110743103/1167829464189779999/1167829464189779999

Acknowledgements

• [X] My issue title is concise and descriptive.
• [X] I have searched the existing issues to make sure this feature has not been requested yet.
• [X] I have provided enough information for the maintainers to understand and evaluate this request.

[homanp]

@p6l-richard What do you think about adding a flag to the datasource in which a user can enable using OCR?