[Feature]: discard soft 404

Feature Summary

if a server provides an *.xml which is actually a 404 html page discard it

What problem does this feature solve?

Version: 2.3.2 Git Commit: Unknown (built outside VCS) Build Date: Unknown (built outside VCS) Go version: go1.26.4 Postgres version: 15.18

Reproduce

1. Add the following url: https://manuelcortez.net/2023/08/creating-vm-libvirt-linux-windows/
2. Miniflux finds many feeds, e.g. https://manuelcortez.net/atom.xml, which is in reality a 404 html page.

 $ curl -si https://manuelcortez.net/atom.xml | head -n 1
HTTP/2 200

 $ curl -si https://manuelcortez.net/atom.xml | grep -o 404
404

$ curl -si https://manuelcortez.net/atom.xml | grep -o "rss <version" || echo "Not found"
Not found

The server is built with HUGO, so the real feed is at https://manuelcortez.net/index.xml, which miniflux finds as well.

It is probably a misconfiguration on the website side, but it would be nice if miniflux would discard them since the content is not an xml/rss.

I'm not sure if miniflux supports html feeds but I was guessing only xml/rss.

Proposed Solution

Check the first bytes of a possible feed to see if it starts with "<rss version" (or similar feed signatures). Something similar to magic bytes for file signature.

Alternatives Considered

No response

Additional Context

No response

Checklist

• I have searched existing issues to ensure this feature hasn't been requested before.
• I understand that feature requests are not guaranteed to be implemented.
• I agree to follow the project's contribution guidelines.