Currently you cant customize host level alerts at the host level only at the windows cluster level.

A few scenarios where this is needed :-

Always-on cluster where one node is for the application and you have a read-only secondary for reporting. I need to be alerted for the application node for high CPU, but the report server is usually often hits 100% CPU during business hours.

Multi node regular failover cluster with different application owners. 3 nodes, app1 lives on node1, app2 lives on node2, node 3 is a passive node for both. Alerts for app1 need to go to team1 alerts for app2 need to go to team2.