feat(instances): the VM lists follow changes by themselves

A VM list showed what it read when it was loaded: a VM stopped from
inside its guest or with virsh, or made or removed elsewhere, showed only
after a reload, unlike the VM page, which reloads when its VM's state
changes. The lists of all VMs and of one host's VMs now poll the states
of their VMs, at once and then every 5 s as the VM page does, and reload
when an answer differs from the one before. The states come from one
libvirt call per host; a user who does not see hosts gets only their own
VMs. The page in use (a dialog, a field, an unsaved XML edit, a job
outcome shown) holds the reload until it is free; that check is shared
with the job panel.

instances/utils.py

@@ -338,6 +338,25 @@ def record_migration(instance, new_compute):
             instance.save()
 
 
+def vm_states(user, computes):
+    """{compute pk: {"ok": True, "vms": {uuid: [name, state]}}, or {"ok": False}
+    for a host that cannot be read}: what the VM lists poll, read from libvirt;
+    for a user who does not see hosts only their own VMs (a VM is its host and
+    UUID: one UUID may be defined on two hosts)."""
+    show_hosts = user.is_superuser or user.has_perm("instances.view_instances")
+    own = None if show_hosts else set(
+        Instance.objects.filter(userinstance__user=user).values_list("compute_id", "uuid"))
+    result = {}
+    for compute in computes:
+        try:
+            vms = compute.proxy.get_domain_states() if compute.status is True else None
+        except libvirtError:
+            vms = None
+        result[str(compute.pk)] = {"ok": False} if vms is None else {
+            "ok": True, "vms": {u: s for u, s in vms.items() if own is None or (compute.pk, u) in own}}
+    return result
+
+
 def migration_target(compute):
     """The destination for the log: its host name, and the migration address
     when one is set and differs (else the migration goes to the host name)."""