[FEATURE REQUEST]scipy-compatibility of cho_solve

[v923z]

@vikas-udupa

Vikas, according to the documentation, https://docs.scipy.org/doc/scipy/reference/generated/scipy.linalg.cho_solve.html, the first argument of cho_solve is a tuple, while #391 implements it with a single matrix. I overlooked the issue at the time, but I think I would like to have scipy-compatibility. Do you want to look into this, or should I?

[vikas-udupa]

For the moment, I implemented cho_solve such that it is compatible with numpy.linalg.cholesky function that has already been implemented. But in the long run, it is right to follow what is given in the documentation, just like you mentioned.
However, I am unable to work on it at the moment due to other commitments. It would be great if you could go ahead and fix this. If not, I can fix this after a couple of weeks.

[v923z]

@vikas-udupa OK, thanks for the feedback! I might take a stab at it.