Cleanup policy for `shedlock`

[sahibamittal]

Current Behavior

shedlock table holds the logs for locks acquired so far by different tasks/processes. But this data is not required after the task is finished. It is resulting into thousands of tangling records in the table.

Proposed Behavior

Create a cleanup policy for records in the table.
Retention period TBD, or delete records whose lock_until is in past.

Checklist

• I have read and understand the contributing guidelines
• I have checked the existing issues for whether this enhancement was already requested

[nscuro]

Do we still have tasks that create locks with dynamic names? We used shedlock for BOM upload processing a while ago, but have since moved to Postgres advisory locks.

I think if the lock names are static and those static names are small in number, the work to enforce retention likely outweighs the benefits.