Generalize for hermitian PD (complex)

[chriscoey]

Is this just trivial to do?

[andreasnoack]

Yes and no. I think much of it should be as simple as loosening the signatures but there might be places where the package makes use of the elements being real in non-obvious ways so I suspect you'd have to extend the testing a lot to be sure that this actually works.

[chriscoey]

Makes sense, thanks!