RFC: implement "catch signal"

[Tom Tromey]

>>>>> "Jan" == Jan Kratochvil <jan.kratochvil@redhat.com> writes:

Jan> I proposed the scheme above so that 'what' can remain still a
Jan> string while it also differentiates the various sub-kinds of the
Jan> 'catchpoint' type.

Yeah, but why?
The "type" field is clearly intended for this.
It seems strange to overload an existing field and then add new fields.
Furthermore I don't think that the current situation was done
intentionally.  It must be an oversight.

Tom