Expose a readiness check endpoint to allow validating that Postgres is still up

[gastonfournier]

The readiness check should be an aggregate of your ability to serve traffic and the health of the infra you depend on. When postgres is unavailable on startup, Unleash exits correctly, but if it becomes unavailable after startup, the health check continues to return healthy.

Have the ability to opt in on a more aggressive readiness check that also validates that the underlying infrastructure on which Unleash depends, is up.

Originally posted by @msanssouci in #10670 (reply in thread)