Curly brackets may be used for sets but they are also used for different cases of functions, or as a more succinct and precise way of expressing sentences about multiple things than writing “foo, respectively bar” all over the place, or for Stirling numbers of the second kind, or for whatever other things an author might want.
Other “structure making” notations like group presentations use angle brackets. Ideals are sometimes written with their generators in angle brackets, sometimes with round brackets. Similarly tuples or sequences may be written with either.
Matrices are sometimes written with square brackets and sometimes round, and never with commas. Vectors typically are written with round brackets but may be square. Row vectors sometimes have commas.
Function application is sometimes written with round brackets. Arguments are sometimes separated by commas, sometimes by semicolons, sometimes with one argument as a subscript or implicit. Application often omits brackets altogether. Sometimes the function goes on the left, sometimes it goes on the right.
Now consider the few features which I find are relatively common to different mathematics notations are context, alignment and structure in 2 dimensions, juxtaposition, implicitness, and terseness. I think I see basically none of these in the common C/Rust style syntax you claim to be so rooted in mathematical notation, and indeed I would guess that you would oppose them.
I don’t buy your arguments at all.
