Fixing renaming FKs by adding with_lock_retries

What does this MR do and why?

Fixing renaming FKs by adding with_lock_retries

Changelog: other

https://app.incident.io/gitlab/incidents/3776

The migration was introduced in !203323 (merged)

It's worth noting that the incident on Production was resolved by just retrying the migration. So this is also to avoid the problem on self-managed.

CC: @stanhu

MR acceptance checklist

Evaluate this MR against the MR acceptance checklist. It helps you analyze changes to reduce risks in quality, performance, reliability, security, and maintainability.