I'm still trying to understand what went wrong though. Putting a finer point on my question: Does pg_upgrade mess up disabled subscriptions?
yes, whether they're disabled or not. As far as I know it's impossible to reliably pg_upgrade a node that has subscriptions and eventually resume logical replication.
It's possible to make it work with some efforts in some basic configurations and / or if no changes happen on the publications, but it's up to you trying to find out if your specific scenario can work as it's not documented. It's also impossible to check whether some incompatible events happened on any of the publisher nodes so you have to make sure that you have total control and knowledge of all the activity that happens on the publisher nodes during the upgrade.