[PATCH] Use the inf-norm rather than 2-norm in issymmetric; add ishermitian.

[Jason Riedy]

And Rolf Fabian writes:
> After the current change of 'issymmetric'
> complex hermitian matrices are not any more considered
> to be 'symmetric' as before and many existing programs
> fail or return weird results.

I would argue that many existing programs *already* return weird
results with no warning.  Complex symmetric is a very, very
different beast than complex Hermitian.  I'd be more comfortable
seeing issymmetric removed altogether than changed to be
isnotreallysymmetricbutwegottakeepthis.

And yes, there are a zillion other classes.  My main point was to
support a common structure while normalizing a bit of terminology
between Octave, LAPACK, and linear algebra.

Jason