The lifecycle-coupled core of one microVM: the jailed firecracker daemon and its controller, restarted as a pair. Isolated from the API client (Hyper.Node.FireVMM.Client) so the only order-sensitive relationship in the VM tree lives in this two-child supervisor.

1. Hyper.Node.FireVMM.Daemon - the jailer OS process (under MuonTrap). MUST be the first child: it must be (re)started before the controller so the API socket is coming up by the time the controller probes.
2. Hyper.Node.FireVMM.State - the :gen_statem controller; drives the boot protocol (await API -> stage -> configure -> run) against that daemon.

:one_for_all, daemon first: a crash of either child takes both down and restarts the pair. So:

• controller crash -> daemon also discarded; no orphaned VM, fresh cold boot.
• firecracker crash -> the Daemon child exits; both restart; Daemon resets the stale jail and relaunches, and the fresh controller cold-boots.

Daemon guarantees firecracker is dead on teardown via the helper's cgroup.kill (MuonTrap's port-close kill misses the setsid'd firecracker), so no firecracker process outlives a graceful supervisor shutdown.