I guess you could stretch and say he means "myth" like you might use "lore" or "mythology" as in the collected beliefs of a group or people, but that's stretching it, if you ask me.
I'm not a DBA, but I really like reading articles about common pitfalls and mistakes, to broaden my knowledge base. Something like "'SELECT *' and your ORM might be using it for you" would have been a much more informative title to me, without dancing around the clearly intended use of the word "myth," and I personally would have still read it, without expecting content completely at odds with what I would say is the natural interpretation of the title.
His argument seems to be that it's a "myth" because you could do the same thing by explicitly listing out the columns. Which in no way makes the former a myth, and it is just pure clickbait, as you said.
It gets much worse than just single table grab-alls, though. I once worked at a shop where they insisted upon prolifically using "select just about everything from every relational table, comprehensively joined to the n-th level" views as some sort of confused notion about code reuse, such that everything would then select from these master views. The end result was that it was incredibly frustrating trying to resolve performance problems, because literally everything was a performance problem. It was one of those instances where I shook my fists at every naive developer railing about purported premature optimizations.
"Don't go out of your way to optimize prematurely" does not mean "architect your software in the opposite direction to industry common sense". "We will optimize later if needed" does not work at architecture level. That's why you prototype before committing yourself.
"Don't go out of your way to optimize prematurely" does not mean
It doesn't mean anything, because both optimize and premature are in the opinion of the viewer. There is no consensus on this at all, but it has been my experience that when those famous words are uttered, badness is about to occur.
In conclusion: Clickbait for DBAs