Whilst testing cruise-control integration with managed kafka (see https://github.com/bf2fc6cc711aee1a0c2a/kas-fleetshard/pull/698), we've run into what looks like a Strimzi defect.  Fleetshard is configuring 'cruise.control.metrics.topic.min.insync.replicas' to 2 in order to comply with Managed Kafka's CreateTopicPolicy validation rules (minISR).  However, once that is done, Strimzi internal validation fails like so:

2022-04-21 10:38:22 WARN  AbstractOperator:532 - Reconciliation #21(timer) Kafka(kafka-c9g2o4iob8uvjht9r7rg/penguin): Failed to reconcile java.lang.NumberFormatException: null     at java.lang.Integer.parseInt(Integer.java:614) ~[?:?]     at java.lang.Integer.parseInt(Integer.java:770) ~[?:?]     at io.strimzi.operator.cluster.model.KafkaCluster.configureCruiseControlMetrics(KafkaCluster.java:609) ~[io.strimzi.cluster-operator-0.26.0.managedsvc-redhat-00010.jar:0.26.0.managedsvc-redhat-00010]     at io.strimzi.operator.cluster.model.KafkaCluster.fromCrd(KafkaCluster.java:425) ~[io.strimzi.cluster-operator-0.26.0.managedsvc-redhat-00010.jar:0.26.0.managedsvc-redhat-00010]     at io.strimzi.operator.cluster.operator.assembly.KafkaAssemblyOperator$ReconciliationState.lambda$getKafkaClusterDescription$56(KafkaAssemblyOperator.java:1481) ~[io.strimzi.cluster-operator-0.26.0.managedsvc-redhat-00010.jar:0.26.0.managedsvc-redhat-00010]