Fix 220

[cvaliente]

@holdenk taking up #228 again.
The full outer join does not care about ordering or partitioning and works on distributed, larger datasets.
I also included an optional skip of the schema equality. One common use case was to load input and expected data from a csv and compare the transformed input to expected output. when loading a csv Spark automatically makes all fields nullable regardless of any schema. the output of the tested functionality could have non-nullable fields (e.g. groupby().count() results in a non-nullable count column)
For unit tests it's often just interesting whether or not the data matches, not if spark inferred the schema correctly

[cvaliente]

pipeline failure seems to be unrelated.

[codecov-io]

Codecov Report

Merging #308 into master will increase coverage by 7.52%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##           master     #308      +/-   ##
==========================================
+ Coverage   78.35%   85.87%   +7.52%     
==========================================
  Files          45       11      -34     
  Lines         388      354      -34     
  Branches       34       34              
==========================================
  Hits          304      304              
+ Misses         63       29      -34     
  Partials       21       21

Flag	Coverage Δ
#python	85.87% <ø> (ø)	⬆️
#scala	?

Impacted Files	Coverage Δ
...scala/com/holdenkarau/spark/testing/Prettify.scala
...ldenkarau/spark/testing/StreamingSuiteCommon.scala
.../holdenkarau/spark/testing/MLUserDefinedType.scala
...oldenkarau/spark/testing/StreamingActionBase.scala
...om/holdenkarau/spark/testing/TestInputStream.scala
...a/com/holdenkarau/spark/testing/RDDGenerator.scala
...la/com/holdenkarau/spark/testing/YARNCluster.scala
...enkarau/spark/testing/JavaDataFrameSuiteBase.scala
...holdenkarau/spark/testing/DataframeGenerator.scala
...nkarau/spark/testing/StructuredStreamingBase.scala
... and 23 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 79eef40...c9634c0. Read the comment docs.

[holdenk]

Sorry for my delay in merging this I was in a motorcycle crash. Looks good to me.