fix(discourse): Remove early break that was limiting topics to indexing batch size

[suvodhoy]

Description

The Discourse connector was incorrectly breaking after processing the first batch of topics (16) even though the API returns 30 topics per page. This caused the connector to miss processing the remaining topics from each page.

How Has This Been Tested?

Noticed a discrepancy in the document count indexed when trying to sync my own discourse setup. I tested by making the changes and building the background service locally and running the sync again.

Backporting (check the box to trigger backport action)

Note: You have to check that the action passes, otherwise resolve the conflicts manually and tag the patches.

• This PR should be backported (make sure to check that the backport attempt succeeds)
• [Optional] Override Linear Check

[vercel]

@suvodhoy is attempting to deploy a commit to the Danswer Team on Vercel.

A member of the Team first needs to authorize it.

[greptile-apps]

PR Summary

Fixed a critical performance bug in the Discourse connector where document indexing was prematurely stopping after the first batch, causing incomplete topic synchronization.

• Bug fix in backend/onyx/connectors/discourse/connector.py removes incorrect break statement that limited topic processing to first batch size (16) instead of full API page size (30)
• Improves separation of concerns between topic collection and document batching logic
• Ensures complete processing of all available topics from each API page
• Aligns with principle to fail loudly by fixing silent data loss issue

_{1 file reviewed, no comments
Edit PR Review Bot Settings | Greptile}

[Weves]

Thanks for this fix!