Before, we used a procd netdev trigger, but that fired *way* too often (maybe
once per router announcement received?), causing high load for nothing.
This basically reverts to the behaviour before commit d8bb978, only that the
init script still handles argument collection.
Because we unconditionally appended `-i br-client` to the command line of
respondd, it wasn't restarted when br-client changed state. Now, we use a
jsonfilter expression on the network.interface dump data, similar to how the
other interface names are generated, and only add the interface to the argument
list if it is up.
Instead of starting gluon-respondd from a hotplug handler, add a proper
init script. The new init script has a restart_if_running argument which
is now used by the hotplug handler.
