Fixed stuck reconciliation on failed request rebalance

[ppatierno]

Type of change

• Bugfix

Description

This PR fixes #10631.

What's happening right now is that after issuing a proposal request again (when the user task doesn't exist, i.e. CC rolled), we are not handling the failure (in this case the "brokers to be removed don't exist anymore", they were already scaled) so the lock on the KafkaRebalance is never released and the reconciliation on it never ends (a failure log shows a never ending "Reconciliation in progress" every periodic timer call).
This fix handles the failure scenario allowing the KafkaRebalance to move in NotReady state which means:

• for a manual rebalance, providing the user a way to understand what happened and behave accordingly
• for auto-rebalancing, deleting the KafkaRebalance which is not needed anymore because the rebalancing was done and nodes scaled down.

Checklist

• Write tests
• Make sure all tests pass
• Try your changes from Pod inside your Kubernetes and OpenShift cluster, not just locally
• Reference relevant issue(s) and close them after merging

[ppatierno]

This PR was also tested through the @see-quick STs which were failing exactly on this #10666

[ppatierno]

/azp run regression

[azure-pipelines]

Azure Pipelines successfully started running 1 pipeline(s).

[scholzj]

LGTM. Thanks.