As per discussion on Zulip, this is not a bug, since:

1. JPA query hints are optional, and
2. Postgres itself does not support statement-level lock timeouts.

Again, as per the discussion there, we think it's in principle possible to implement this using set statements, but that’s not, IMO, a terribly natural implementation, involving multiple round trips to the database for a single query.

So I would sat that, aside from not being a bug, this is also a very low-priority improvement.