Apparently enabling Hyper-V polling has a side effect of kicking the node out of a dynamically queried group and changing the way the node behaves and is managed in Solarwinds.
For instance, you can't edit node properties from the node page. It no longer being a member of a group affects the group's health status as well: the node can be down, the group still up.
Perhaps SW developers were thinking of VMWare when they were baking Hyper-V polling. Virtualization is the only thing VMWare does. Hyper-V: nuh-uh. I can quietly run a bunch of test VMs on a side in Hyper-V with other roles and apps still being mission critical. So perhaps re-think that a little bit? Until then, no Hyper-V management for me.