[Task]: Write promoted group directly to new models for multiple promoted groups

[chrstinalin]

What happened?

With #15350, promoted group information is being read from new database models while writes are synced from the promoted/signals.py file (introduced with #15335). We should transition to writing directly to the new database models rather than relying on writes to the legacy classes.

What did you expect to happen?

1. Promoted group information is written directly to the new database models.
2. The use of PromotedClass is retired.

Is there an existing issue for this?

• [x] I have searched the existing issues

┆Issue is synchronized with this Jira Task