if you're hosting services then I assume you work or aspire to work in IT so you should probably be wanting to learn it even if there isn't a golden carrot to tempt you.
What I'm talking about here is specifically past the learning stage. It's important to learn how to jump start a car, but once that's done I'm not going to do it unless I have a reason to.
The reason why I would go to VPNs or CF tunnel style forwarding services (there are plenty of alternatives like Ngrok) is because I can't use IPv6 as a replacement for them, namely to enable connecting from networks which do not have IPv6, which is a rather common requirement.
Now, once that's done, setting up IPv6 doesn't let me do anything more, other than in very specific/contrived situations.
If I don't stand to gain anything from it (once again, approaching from an already learned context), I'd tend to just disable it to both save myself the effort, and reduce the attack surface.
Also, if I want to hide by my home IP or get some DDoS protection (much easier to rotate VPNs than force an IP change in some cases), I still have to setup a VPN and NAT (or tunnel) anyway.
What's the "effort", in your case?
Other than the those you mentioned, like assigning, documenting, and designing a new set of addresses, even assuming no compatibility issues or the setup itself running into issues:
- The layout is different, If your IPv4 has NAT and IPv6 is different, it's not as simple as tacking on 1 more address
- Binding to specific addresses for security? Remember to update those too
- Same for the firewall rules and NAT (if you wish to hide your IP)
- Something not working? Need to test if the issue is with IP 4, 6, or both. Also time spent testing that it actually works on 6, and not just falling back to 4
Is it much more effort? Not really. But it's against the ratio of zero benefit, since whatever solution I use to get stuff working for IPv4 exclusive networks already solves the issue.