#4646 minor hstep problems and missing features

[Barton Willis]

This is OK

(%i1) hstep(x^2+1);
(%o1)                                  1

But this is not OK--it should be a hstep nounform:

(%i2) declare(x,complex)$
(%i3) hstep(x^2+1);
(%o3)                                  1

Not wrong, but pz is better:

(%i1) sign(hstep(x));
(%o1)                                 pnz

And we should give hstep an antiderivative property:

(%i3) integrate(hstep(x),x);

(%o3) 'integrate(hstep(x),x)

I have a putative fix for these deficiencies.

[Barton Willis]

Fixed by Commit [a77814] . Closing this ticket.