Bugfix/wald

[oleksandr-pavlyk]

@samir-nasibli

This fixes incorrect sampling in Wald, a.k.a Inverse Gaussian distribution.

(mkl_random_dev) [16:32:33 lin03 mkl_random]$ ipython
Python 3.6.8 |Intel Corporation| (default, Apr 12 2019, 17:34:58)
Type 'copyright', 'credits' or 'license' for more information
IPython 6.3.1 -- An enhanced Interactive Python. Type '?' for help.

In [1]: import mkl_random, numpy as np

In [2]: wald_sample = mkl_random.wald(mean=14, scale=15, size=5*10**6)

In [3]: np.mean(wald_sample)
Out[3]: 14.007203796988938

In [4]: np.var(wald_sample)
Out[4]: 183.2745041415038

In [5]: np.mean(np.logical_and(13 < wald_sample, wald_sample < 17))
Out[5]: 0.1069688

All sample expectations agree with population expectations.