Implement (DataFrame|Series).plot.kde with Plotly

[HyukjinKwon]

This PR implements (DataFrame|Series).plot.kde with Plotly:

Can be tested here:

http://mybinder.org/v2/gh/hyukjinkwon/koalas/plotly-kde?filepath=docs%2Fsource%2Fgetting_started%2F10min.ipynb

matplotlib:

import databricks.koalas as ks
ks.options.plotting.backend = "matplotlib"
kdf = ks.DataFrame({
    'a': [1, 2, 2.5, 3, 3.5, 4, 5],
    'b': [1, 2, 3, 4, 5, 6, 7],
    'c': [0.5, 1, 1.5, 2, 2.5, 3, 3.5]})
kdf.plot.kde(bw_method=3).figure.savefig("fig1_mat.png")
kdf.a.plot.kde(bw_method=3).figure.savefig("fig2_mat.png")

plotly

import databricks.koalas as ks
ks.options.plotting.backend = "plotly"
kdf = ks.DataFrame({
    'a': [1, 2, 2.5, 3, 3.5, 4, 5],
    'b': [1, 2, 3, 4, 5, 6, 7],
    'c': [0.5, 1, 1.5, 2, 2.5, 3, 3.5]})
kdf.plot.kde(bw_method=3).write_image("fig1.png")
kdf.a.plot.kde(bw_method=3).write_image("fig2.png")

[codecov-io]

Codecov Report

Merging #2028 (7db02e4) into master (5e28195) will decrease coverage by 4.02%.
The diff coverage is 32.69%.

@@            Coverage Diff             @@
##           master    #2028      +/-   ##
==========================================
- Coverage   94.59%   90.57%   -4.03%     
==========================================
  Files          50       50              
  Lines       11277    11210      -67     
==========================================
- Hits        10668    10153     -515     
- Misses        609     1057     +448     

Impacted Files	Coverage Δ
databricks/koalas/plot/plotly.py	9.47% <5.26%> (-86.58%)	⬇️
...bricks/koalas/tests/plot/test_frame_plot_plotly.py	23.33% <14.28%> (-76.67%)	⬇️
...ricks/koalas/tests/plot/test_series_plot_plotly.py	26.92% <14.28%> (-71.14%)	⬇️
databricks/koalas/plot/core.py	91.49% <72.22%> (-1.04%)	⬇️
databricks/koalas/plot/matplotlib.py	91.27% <100.00%> (-0.58%)	⬇️
databricks/koalas/usage_logging/__init__.py	25.66% <0.00%> (-66.65%)	⬇️
databricks/koalas/usage_logging/usage_logger.py	47.82% <0.00%> (-52.18%)	⬇️
... and 25 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 5e28195...7db02e4. Read the comment docs.

[HyukjinKwon]

This was a bug BTW. it had to be max_val - min_val

[HyukjinKwon]

Added a logic to get the max/min from multiple columns

[ueshin]

LGTM.