One easy way to see that "linear" does not mean "line-like" is looking at matrix transforms: All matrix transforms are linear. And we can do a lot of stuff easily with those, like rotating, distorting, even perspective (with w-normalization).
All matrix transforms are linear. And few people understand matrix transforms. The term itself "linear" transform stems more from the contrast to nonlinear equations, but it still "feels" like a bit of a misnomer. Perhaps it should be called "proportional" or "relatable" transform.
Ehh, then again, it doesn't really matter, people who don't understand it still won't with a nomenclature change.