Feature request: support for binding remote services to subnet addresses

[dimka511]

Hello I’d like to suggest a small but useful enhancement:
please add the ability to use addresses from the node’s subnet (obtained via -subnet) when defining remote endpoints.

Use case

Currently, when exposing local services via Yggdrasil, we can use:

./yggstack -useconffile /path/to/yggdrasil.conf -remote-tcp 80:127.0.0.1:8080
./yggstack -useconffile /path/to/yggdrasil.conf -remote-udp 53:127.0.0.1:53

It would be very helpful if the syntax allowed specifying either:

a full IPv6 address from the subnet, e.g.

-remote-tcp [3xx:xx:xx:xx::1]:80:127.0.0.1:8081
-remote-tcp [3xx:xx:xx:xx::2]:80:127.0.0.1:8082

or a suffix (for example, ::1, ::2, etc.) relative to the node’s subnet prefix.

This would allow hosting multiple services on different IPv6 addresses within the same subnet, all through a single running instance of yggstack.

Additional idea

It might also make sense to allow specifying a subnet address (or suffix) for the SOCKS proxy binding:

./yggstack -useconffile /path/to/yggdrasil.conf -sockssubnet [3xx:xx:xx:xx::10] 127.0.0.1:1080

Even though multiple SOCKS proxies are probably a rare use case, having consistency in how subnet addresses can be used would be great.