A common case is for typeahead completion. You have a requst to the sever in flight to get typeaheads but the user has entered more characters and triggered another request. Now you would like to ignore the old one. Since async calls are not guaranteed to arrive in the same order they were sent, it would be nice to cancel the old one instead of maintaining logic to ignore it.
I think you nailed it about his drive for successful serial entrepreneurship. Slide was sold for an ok price but was shut off. I don't know how Levchin fared at Google but I assume he probably didn't like working for someone else. I think he chose an area where it was almost guaranteed they would make a lot of money / perceived success. Affirm is definitely that, they will make a lot of money, but it is not far off from payday loans and rent-to-own furniture, no matter how good the PR is.
Most charting frameworks tend to have default configurations which demonstrate more features than necessary as an advertisement of their competence for potential users, who are likely to copy default configuration and also take demo charts as cues on what charts should look like. It isn't the only dynamic which causes charts to tend to have superfluous or untuned features but it has some effect.