Make it compatible with numpy > 1.16

[jfosorio]

In version of numpy higher than 1.16 the third argument of linspace is not be interpreted as integer and would produce and error.

https://docs.scipy.org/doc/numpy/release.html?highlight=histogram

Then matrix d_mat need to be initialized as a integer type.

[alexpacini]

Is it like #27 ?

[jfosorio]

Yes, I think is the same but the solution is slightly different.
Here I do it while creating the matrix:
d_mat = np.ones((len_x, len_y, 2), dtype=np.int)
#27 do it just before the information is used in:
for xs in np.linspace(x0, x1, int(x_div + 1))[1:]:

[afiskon]

I can confirm that this patch fixes the compatibility issue with numpy > 1.16. Hope to see it in the upstream!

[DocEpsilon]

This one-line fix is an open issue for a year now. Maybe add some contributors? :-)
A lot of people forked obviously because of this.

[afiskon]

@vMeijin this is a major issue since numpy==1.6 can't be compiled on some modern systems anymore with recent Python versions (I'm using MacOS). Could you please just press the "Merge" button? :)

[SaladePengouin]

@rpcope1 maybe could merge it? Thanks.

[rpcope1]

@SaladePengouin I wish I could, I don't have the ability to merge. I haven't thought about library for almost a decade now, but this sounds like a real challenge. I don't know if there's a preferred fork or something, otherwise I might be able to take a stab at it.