Python client does not reconnect if initial connection fails with ENETUNREACH

ENETUNREACH can happen when target network is unreachable for example when the network stack was not fully initialized yet or when a network is not connected temporarily, etc.
Since this type of error can very much be transient - I'd expect proton to try (re)connect as usual (just like it does for EHOSTUNREACH).
This is on recent Fedora.

Easy ways to reproduce include disabling network connection and running proton trying to connect to some non-link-local address.