Add diff in Groupby

[RainFung]

This PR adds rank GroupBy (both SeriesGroupBy and DataFrameGroupBy) by using existing diff logic in Series.

[codecov-io]

Codecov Report

Merging #622 into master will increase coverage by 0.08%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master     #622      +/-   ##
==========================================
+ Coverage   92.79%   92.87%   +0.08%     
==========================================
  Files          31       31              
  Lines        5008     5023      +15     
==========================================
+ Hits         4647     4665      +18     
+ Misses        361      358       -3

Impacted Files	Coverage Δ
databricks/koalas/missing/groupby.py	100% <ø> (ø)	⬆️
databricks/koalas/series.py	92.73% <100%> (+0.02%)	⬆️
databricks/koalas/groupby.py	80.6% <100%> (+0.92%)	⬆️
databricks/koalas/frame.py	94.63% <0%> (+0.07%)	⬆️
databricks/conftest.py	97.67% <0%> (+2.32%)	⬆️
databricks/koalas/__init__.py	84.21% <0%> (+2.63%)	⬆️

---

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 e141bf7...06c7f56. Read the comment docs.

[HyukjinKwon]

Looks fine in general.

[softagram-bot]

Softagram Impact Report for pull/622 (head commit: 06c7f56)

⭐ Change Overview

(Open in Softagram Desktop for full details)

📄 Full report

• Permalink: Full report for pull/622

Give feedback on this report to [email protected]

[HyukjinKwon]

Thanks, @RainFung for working on this.

[ppakawatk]

Hi,

I'm trying to convert my Pandas script to Koalas script, and found this error when trying to do GroupBy.diff with datetime column.

Here's the code to reproduce the case:
df = ks.DataFrame({'a': [1, 2, 3, 4, 5, 6], 'b': [1, 1, 2, 3, 5, 8], 'c': [dt.date(2019, 1, 1), dt.date(2019, 1, 3), dt.date(2019, 1, 10), dt.date(2019, 1, 11), dt.date(2019, 1, 20), dt.date(2019, 1, 21)]}, columns=['a', 'b', 'c'])

df.groupby(['b'])['c'].diff()

Referring to https://koalas.readthedocs.io/en/latest/reference/api/databricks.koalas.groupby.GroupBy.diff.html
I assume that right now it doesn't support datetime column when diff?

PS. here is the Pandas script which is working properly.
df = pd.DataFrame({'a': [1, 2, 3, 4, 5, 6], 'b': [1, 1, 2, 3, 5, 8], 'c': [dt.date(2019, 1, 1), dt.date(2019, 1, 3), dt.date(2019, 1, 10), dt.date(2019, 1, 11), dt.date(2019, 1, 20), dt.date(2019, 1, 21)]}, columns=['a', 'b', 'c'])

df.groupby(['b'])['c'].diff()

[HyukjinKwon]

Can you open an issue with runnable examples? Seems like it's also related with Calender internal type support (SPARK-24695)

[ppakawatk]

Can you open an issue with runnable examples? Seems like it's also related with Calender internal type support (SPARK-24695)

Yes, sir.
#965