Fix DataFrame.replace with NaN/None values #1907

[LucasG0]

Hi, this PR would close #1907.

[codecov-io]

Codecov Report

Merging #1962 (7acdba6) into master (02133a8) will decrease coverage by 3.57%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master    #1962      +/-   ##
==========================================
- Coverage   94.64%   91.06%   -3.58%     
==========================================
  Files          49       49              
  Lines       10861    10760     -101     
==========================================
- Hits        10279     9799     -480     
- Misses        582      961     +379     

Impacted Files	Coverage Δ
databricks/koalas/series.py	95.71% <100.00%> (-1.17%)	⬇️
...bricks/koalas/tests/plot/test_frame_plot_plotly.py	26.58% <0.00%> (-73.42%)	⬇️
databricks/koalas/usage_logging/__init__.py	25.66% <0.00%> (-66.65%)	⬇️
...ricks/koalas/tests/plot/test_series_plot_plotly.py	37.50% <0.00%> (-58.80%)	⬇️
databricks/koalas/usage_logging/usage_logger.py	47.82% <0.00%> (-52.18%)	⬇️
databricks/koalas/__init__.py	73.43% <0.00%> (-17.19%)	⬇️
databricks/conftest.py	88.13% <0.00%> (-11.87%)	⬇️
databricks/koalas/spark/functions.py	88.88% <0.00%> (-7.41%)	⬇️
databricks/koalas/accessors.py	86.43% <0.00%> (-7.04%)	⬇️
databricks/koalas/namespace.py	78.12% <0.00%> (-6.08%)	⬇️
... and 20 more

---

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 02133a8...7acdba6. Read the comment docs.

[itholic]

Otherwise, looks pretty fine to me.

[itholic]

Seems good to me. Let's wait a while for reviews from others :)

[HyukjinKwon]

Thanks @LucasG0. Merged.

[itholic]

Thanks, @LucasG0 !! :D

[LucasG0]

Thanks. :)