[Operator] Zero controller execute can hang indefinitely if Zero Pod is not immediately ready

The zero_controller execute method makes the assumption that a reconciliation event will be triggered once the Zero Pod becomes ready, however in some circumstances (consistently on CI) this never happens. Therefore, we should explicitly requeue the reconcilliation when the zero pod is not available.

Furthermore, we should add an additional check to the Backup and Restore Exec methods to ensure that resources are only created on the server if they don't already exist.