This also completely ignores the concept of pruning old transactions, which can be done if need be.
Is there a consensus that pruning is actually possible and doable?
Yes. Transactions which have had all outputs spent and have had ? transactions past them are no longer needed to be stored.
And you can apply this to cryptonote blockchain as well? Just making sure we were talking about the same thing.
I don't specifically know enough about it to be sure, however in an extreme hypothetical where cryptonote became the dominant cryptocurrency, beating out even bitcoin, if it was to become an issue, there would be a solution. There are very few unsolvable problems out there, just a lot of problems that don't have enough benefit to get enough smart people working on a solution to solve them.
For me, the fact that all the infrastructure, coding, and development for bitcoin and related currencies is entirely incompatible with it makes cryptonote a nonstarter for me anyway though. So in my mind blockchain bloat is just FUD, because that is a problem that could be solved by one person (if it was a problem.) Brand new integration with hundreds of services, etc, takes coordination of many, and that is a flaw, not FUD.