Freeze node discovery until all nodes are ready

[AlbertoSoutullo]

If a simulation is required with discv5, and if we want to run a large amount of nodes, these might be start interacting between each other before prometheus and grafana are ready.

One solution to this problem would be:

• Start each node in its own subnetwork.
• Once all nodes are ready, set up prometheus + grafana
• Then, stablish all nodes in the default network using update service or in the correspondant subnetwork given by the topology.

Note: Obviously, this update will not be done in all nodes at the same time, specially if we have a large number. Just noting this here so it is considered for experiments.