Happened to me with an even bigger instance because of an asshole admin making shit up. A solution might be to divide up the host of the user comments versus the moderator agents versus receiver of the comments. If your host bans you, that’s it, but if the receiver bans you, that only affects their users, and if a moderator agent group bans you, that only bans you from their distribution group of moderator agents but could be read by other groups.
If a community / group-of-moderator-agents-under-a-community-tag-for-a-particular-host bans you, you’d have to find another groups of moderator agents or accept all that are allowed by your host. Accepting all allowed by your host could only realistically exclude the worst offenders - spammers, doxxers, etc - so you’d really be incentivized to find a better block of moderator agents if you want to avoid certain types of comments. People who want to live in a bubble could live in a bubble but people who want to prioritize the greatest participation would try to find the most lenient host and the most lenient moderation agents, at least to their particular sensitivities.
It would be a truer federated model, but this is not lemmy as it is.