In-place computation of derivatives

[ntezak]

Hi there, I saw that there is already some discussion on memory efficiency here and a mention to allow for in-place RHS's.
This would be extremely valuable to me for the purpose of simulating quantum systems (and other high-dimensional problems such as PDEs) and so I was wondering how interested you are in making in-place ODEs the default.
I would be happy to work on this given your input.

[mauro3]

Yes, I agree, eg #72 uses in-place. I think this can be done without disturbing the high level API.