Codechange: [Network] split CloseSocket and CloseConnection more clearly (#9261)

* Codechange: [Network] split CloseSocket and CloseConnection more clearly

- CloseSocket now closes the actual OS socket.
- CloseConnection frees up the resources to just before CloseSocket.
- dtors call CloseSocket / CloseConnection where needed.

Feature: use Happy Eyeballs to make network connections (TCP-only) (#9199)

Hostnames like "content.openttd.org" resolve into multiple IPv4 and IPv6.
It is possible that either of the IPs is not working, either due to
a poorly configured OS (having IPv6 but no valid route), broken network
paths, or a service that is temporary unavailable.

Instead of trying the IPs one by one, waiting for a 3s timeout between
each, be a bit more like browsers, and stack attempts on top of each
other with slight delays. This is called Happy Eyebells.

Initially, try the first IPv6 address. If within 250ms there is no
connection yet, try the first IPv4 address. 250ms later, try the
second IPv6 address, etc, till all addresses are tried.

If any connection is created, abort all the other (pending) connections
and use the one that is created. If all fail 3s after the last connect(),
trigger a timeout for all.

Codechange: use connection_string in favour of NetworkAddress (#9197)

We now resolve the connection_string to a NetworkAddress in a much
later state. This means there are fewer places constructing a NetworkAddress.

The main benefit of this is in later PRs that introduce different types
of NetworkAddresses. Storing this in things like NetworkGameList is
rather complex, especially as NetworkAddress has to be mutable at all
times.

Additionally, the NetworkAddress is a complex object to store simple
information: how to connect to this server.

Remove: ENABLE_NETWORK switch

This switch has been a pain for years. Often disabling broke
compilation, as no developer compiles OpenTTD without, neither do
any of our official binaries.

Additionaly, it has grown so hugely in our codebase, that it
clearly shows that the current solution was a poor one. 350+
instances of "#ifdef ENABLE_NETWORK" were in the code, of which
only ~30 in the networking code itself. The rest were all around
the code to do the right thing, from GUI to NewGRF.

A more proper solution would be to stub all the functions, and
make sure the rest of the code can simply assume network is
available. This was also partially done, and most variables were
correct if networking was disabled. Despite that, often the #ifdefs
were still used.

With the recent removal of DOS, there is also no platform anymore
which we support where networking isn't working out-of-the-box.

All in all, it is time to remove the ENABLE_NETWORK switch. No
replacement is planned, but if you feel we really need this option,
we welcome any Pull Request which implements this in a way that
doesn't crawl through the code like this diff shows we used to.