NumPy deprecated the `interpolation=` argument for the quantile method

[William Bakst]

https://github.com/tensorflow/lattice/blob/master/tensorflow_lattice/python/premade_lib.py#L1474

the interpolation= argument to quantile was renamed to method=, which has additional options.
Users of the modes 'nearest', 'lower', 'higher', or 'midpoint' are encouraged to review the method they used. (Deprecated NumPy 1.22)

Since method= still supports 'nearest' the fix is as simple as replacing interpolation= with method=

Although this only results in a warning, it's a simple fix and would be nice not to have the warning :)