Stephan Rave

> @sdrave I've implemented all changes, thanks for the review! Unfortunately, one test now fails in `pymor.algorithms.eigs` with `IndexError: index 1 is out of bounds for axis 0 with size...

> Would it be ok to catch the IndexError in the tests and skip them for now until this is solved more rigoriously? Let's first see if we can quickly...

I have removed the special case for `A.sup_norm().max() < 1e-16` as I really do not think that we should test for a 'zero' matrix using some arbitrary fixed tolerance. The...

> As I recall, hypothesis runs tests with the zero array. So, if the tests pass, I'm fine with merging. There's ```python assume(len(U) > 1 or not contains_zero_vector(U)) ``` in...

@pmli, if this isn't needed urgently, I would propose to wait until we have decided how `StationaryModel` will look like in the future. Then I would adjust the RB code...

Removing from milestone due to hard freeze.

@peoe , @Jonas-Nicodemus, are you still interested in working on this PR? Otherwise we would close it ..

@pmli, actually, what do we critically need from slycot that is not covered by scipy?

@artpelling, sorry, was on vacation. Will do proper review by the end of the week.

> I dislike having the `complex` (fka `iscomplex`) arguments in the randomized methods. I dislike it as well. > IMO this should depend on the operator. Does anyone know a...