Support Vm-HA across clusters in case of zone-wide scope primary storage

### The required feature described as a wish

As a User/Admin/Operator

I would like to my HA enabled vm's to be restarted by vm ha across cluster in case the primary storage zone-scoped.

The current behaviour looks for hosts which are in Up state in the same cluster.

https://docs.cloudstack.apache.org/en/4.21.0.0/adminguide/reliability.html#ha-enabled-instances

As zone wide storage spans clusters it would be great the vm's gets started across cluster when a host goes down.

Start with the HA-enabled instances section in the linked CloudStack reliability documentation and compare it with the stated behavior of searching only for hosts in the same cluster. The work is done when HA-enabled VMs using zone-wide primary storage can restart on an available host in another cluster after a host failure.

Written by the indexing model from the issue text.