[Feature] Add `Incus` support

[gyptazy]

As PegaProx already supports alternative hypervisors just like Proxmox and XCP-ng (and Sylve which is WIP), it might also make sense to integrate Incus support which could maybe also benefit at a later point from the other integrations when this could also support cold migrations between solutions.

However, the base idea is to integrate the functionality of VM management, node management, cluster management and metrics for the first draft.

[mkellermann97]

Thanks gyptazy — this fits the direction well. The hypervisor layer is already abstracted enough (Proxmox / XCP-ng / ESXi, Sylve in progress) that adding Incus is mostly a new connector implementing the same VM / node / cluster / metrics surface, so a first draft scoped to exactly that — management + metrics, migrations later — is the right cut, and the cold-migration-between-backends angle is a nice longer-term payoff once the connector exists. No timeline promise since this is volunteer time, but it's on the radar, and you know this codebase well enough that a draft PR would land a lot faster than waiting on us. If you (or someone) wants to sponsor the connector work, opencollective.com/pegaprox helps us carve out the hours. — Marcus