> * ascii-integer?
> * ascii-char?
> * ascii-bytevector?
> * ascii-string?

Having the "ascii-" prefix first is also clearer in that these should
accept any object.

For example, (ascii-integer? "foo") => #f should be perfectly valid, as
should (ascii-bytevector? 123.45) => #f.

I guess we could have (bytevector-ascii? ...) that raises an exception
if the argument is not a bytevector, but that seems more complex than
just returning #f for non-bytevectors.