Yep, that solved it. I removed the “localhost” hostname from the second instance (in the ipv6 section of my hosts file), switched back to the old dpkg version of /etc/default/slapd by doing yunohost tools regen-conf slapd --with-diff --force (to be sure I got the correct slapd config from dpkg) and restarted slapd. As you surmised it would, it came up normally.
Updating to the newest yunohost package also went without incident after making the needed changes to the hosts file.
I’ll probably remove the entire ipv6 section soon as I doubt I’ll have ipv6 support on my fiber connection for decades at the rate things are going here. It’s bizarre to me slapd tries to attach to the same socket on the same device and to the same address without complaining (other than dying outright), though. (For context, I have that same hosts file on dozens of devices running different OSes and daemons, and it’s never been a problem before. 
)
It’s not what I would expect a “sane” daemon to do in such a case, in any rate, I’m glad you saw what I had done to mess things up.
Thanks for all your help, @Josue!