download_results_csv uses slugify without null-check on rnd.name

[ayushshukla1807]

Problem

In admin_endpoints.py, download_results_csv calls:

output_name = 'montage_results-%s-%s.csv' % (slugify(rnd.name, ascii=True).decode('ascii'), now)

If rnd.name is None (which is theoretically possible if a round was created with a missing name before validation was enforced), slugify(None, ...) will raise an AttributeError.

The same pattern exists in download_round_entries_csv, which was fixed in #531. This endpoint has the same latent bug.

Suggested Fix

Add a null-guard:

round_name = rnd.name or 'unnamed-round'
output_name = 'montage_results-%s-%s.csv' % (slugify(round_name, ascii=True).decode('ascii'), now)

[Sanika123-art]

I have implemented the null-check fix for rnd.name in download_results_csv and aligned it with the existing pattern used in download_round_entries_csv.

For testing, I explored the database to identify rounds with existing entries and manually set rnd.name = NULL to simulate legacy/invalid data scenarios. I then verified the behavior through the UI by downloading CSVs for the affected rounds.

Results:

• CSV downloads successfully without crash when name is NULL (uses fallback "unnamed-round")
• Normal cases continue to work as expected
• Empty entries case returns appropriate error message

This ensures the endpoint is safe against missing round names and consistent with similar fixes.

[ayushshukla1807]

Hey @Sanika123-art! Thanks for looking into this and taking the time to do the manual database testing—that's fantastic validation for the fallback behavior!

Just a quick heads-up: I actually opened a PR for this exact fix a couple of days ago in #544 (linked above) while I was doing a larger audit of the CSV export endpoints. Since #544 is already wired up and sitting in the review queue, we'll likely just move forward with that one, but I really appreciate you independently confirming that the "unnamed-round" fallback resolves the UI crashes seamlessly!