1. Create new alert query utilities (`alertQueries.ts`, `metricToAlerts.ts`) and observability hooks (`useClusterCNVInstalled`, `useClusterObservabilityDisabled`) to build Prometheus alert queries with cluster filtering and detect clusters with observability disabled

2. Implement `useClusterObservabilityDisabled` hook that checks for `observability=disabled` label on ManagedCluster resources and filters out clusters without observability when querying metrics

3. Update all overview tab metrics (VM Count, vCPU Usage, Memory, Storage), VM Statuses Card, VMs Per Resource Card, Top Consumers Tab, and Migrations Tab to support observability-aware cluster filtering