Supposing most people think like that, it would indeed explain why there are so little of these types of visualizations out there. The effort, as you say, is probably not worth the objective value it would create (which is why I hoped someone else had already done the work).
For me, a visualization of the breakdown of languages by application domain would have immense value, especially for broadening my perspective. As someone who never gets to toy with FPGAs in my day job for example, I would be oblivious to most of the ideas that flow into PLs for FPGAs. Why a visualization? I guess that is just taste. I find visualization a lot easier and faster to digest.
And I bet there are quite a few PLs that I have no idea about that have answers to problems I have encountered before.
And for the second type of visualization, I find those to be quite helpful in understanding and tracing back the roots of PLs. This often makes me comprehend quite quickly why things are done certain ways in different PLs. Especially looking at the main, root influence of a bunch of PLs and understanding the zeitgeist it was created in, what problem it was solving, who created it etc. give me a much deeper understanding.
So I don't think such visualization are useless, but I could probably not justify spending the amount of time it would take to create them either.