Implemented intersection for Index & MultiIndex

[itholic]

This PR proposes the new API Index.intersection() and MultiIndex.intersection().

>>> idx1 = ks.Index([1, 2, 3, 4])
>>> idx2 = ks.Index([3, 4, 5, 6])
>>> idx1.intersection(idx2)
Int64Index([3, 4], dtype='int64')

[ueshin]

Actually I'd separate the function into ones for each class as there are so many if isinstance(self, MultiIndex): .. else: ...

[itholic]

Actually I'd separate the function into ones for each class as there are so many if isinstance(self, MultiIndex): .. else: ...

Sounds good. I'll separate

[itholic]

@ueshin ,

For #1747 (comment) and #1747 (comment),

The comparison between empty MultiIndex seems not work properly for now.

empty_pandas_multi_index = pmidx.intersection(pidx)  # MultiIndex([], )
empty_koalas_multi_index = ks.from_pandas(empty_pandas_multi_index)  # MultiIndex([], )
self.assert_eq(empty_pandas_multi_index, empty_koalas_multi_index)

The test above is failed with following error.

AssertionError: MultiIndex level [0] are different

MultiIndex level [0] classes are not equivalent
[left]:  Index([], dtype='object')
[right]: Float64Index([], dtype='float64')

Left:
MultiIndex([], )
object

Right:
MultiIndex([], )
object

Maybe let me investigate this in the separated PR ?

[codecov-io]

Codecov Report

Merging #1747 into master will increase coverage by 0.00%.
The diff coverage is 94.00%.

@@           Coverage Diff           @@
##           master    #1747   +/-   ##
=======================================
  Coverage   94.19%   94.20%           
=======================================
  Files          40       40           
  Lines        9867     9915   +48     
=======================================
+ Hits         9294     9340   +46     
- Misses        573      575    +2     

Impacted Files	Coverage Δ
databricks/koalas/missing/indexes.py	100.00% <ø> (ø)
databricks/koalas/indexes.py	97.05% <94.00%> (-0.22%)	⬇️
databricks/koalas/spark/accessors.py	94.96% <0.00%> (+0.62%)	⬆️

---

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 62fb01c...7197f62. Read the comment docs.

[HyukjinKwon]

The empty one would likely be an issue from Spark IIRC.