I generally agree. If you're not doing it for money you don't technically need most of these things. But if you see open source as more than “here's the code” some of them matter. Support will find you, via GitHub issues, emails, or DMs. Analytics is really important because it shows whether the software works for people besides you. Without money you usually do not have playtesters or a UX designer, so you get fewer useful bug reports. Frustrated users rarely take the time to write a detailed issue.
Have you looked at something like Latex or Typst? They come with their own layout engine, so potentially less tedious work like specifying exact positions.
It's not a failure, it's big techs greatest feature. Try to imagine the likes of Windows, Office, Apple, AWS, Social Media etc. without lock-in, network effect or all the other names for missing interoperability.
How would profitability fail the company? Too expensive products won't be bought means no profit, but if you can get away with the price, you're not failing, are you?
I don't think this analogy works, the space laser is instant and does not spread to non-targets.
Smoking does reduce the average life span, but not to zero. In the remaining time, healthcare costs are increased on top of anything expensive they'd develop naturally.
Smoking also causes serious diseases in non-smokers and kills 1.3 million non-smokers per year. https://www.who.int/news-room/fact-sheets/detail/tobacco
Contributions to a fork could be done under just the AGPL, without any CLAs (also to the original repo, but those won't be accepted). Then the entire fork is effectively AGPL only.
I don't think any original CLAs would apply to the fork, unless the fork owner is the same legal entity/successor as defined in there.
Same goes for the original authors, they'd need CLAs from all fork contributors.