Levenberg Marquardt Implementation Tests vs SciPy

[ChrisRackauckas]

Are there any tests against the SciPy LM implementation to see if this implementation is as robust as possible (assuming the SciPy implementation is good)? I can't seem to find out how to properly use PyCall to use the SciPy curve_fit function, but I think it would be a necessary test since the current LM implementation is lacking a lot of robustness (and while that's a standard feature of the algorithm, it just seems even pickier than normal)