cum_transform_feature Docstrings

[glentennis]

Rewriting docstrings in cum_transform_feature.py to include a full description and doctest.

[codecov]

Codecov Report

Merging #494 into master will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master     #494   +/-   ##
=======================================
  Coverage   96.09%   96.09%           
=======================================
  Files         108      108           
  Lines        8862     8862           
=======================================
  Hits         8516     8516           
  Misses        346      346

Impacted Files	Coverage Δ
...tools/primitives/standard/cum_transform_feature.py	100% <ø> (ø)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 3bfc61c...a025ae4. Read the comment docs.

[glentennis]

@kmax12 I edited the descriptions and doctests to be more clear about the way NaN values are handled, let me know what you think!

Also, I noticed the way NaN values are handled in CumMean might not be optimal. NaN values in a window are treated as zero, so when given the list [1, None, 1], CumMean returns:

[1, nan, .66]

When perhaps we would expect it to return

[1, nan, 1]

should i open an issue to fix this?

[kmax12]

LGTM