…ion in PySpark

Author: Jeff Zhang <zjffdu@apache.org>

Closes #9791 from zjffdu/SPARK-11804.

JIRA: https://issues.apache.org/jira/browse/SPARK-11322

As reported by JoshRosen in [databricks/spark-redshift/issues/89](https://github.com/databricks/spark-redshift/issues/89#issuecomment-149828308), the exception-masking behavior sometimes makes debugging harder. To deal with this issue, we should keep full stack trace in the captured exception.

Author: Liang-Chi Hsieh <viirya@appier.com>

Closes #9283 from viirya/py-exception-stacktrace.

JIRA: https://issues.apache.org/jira/browse/SPARK-9166

Simply capture and hide `IllegalArgumentException` in Python API.

Author: Liang-Chi Hsieh <viirya@appier.com>

Closes #7497 from viirya/hide_illegalargument and squashes the following commits:

8324dce [Liang-Chi Hsieh] Fix python style.
9ace67d [Liang-Chi Hsieh] Also check exception message.
8b2ce5c [Liang-Chi Hsieh] Merge remote-tracking branch 'upstream/master' into hide_illegalargument
7be016a [Liang-Chi Hsieh] Capture and hide IllegalArgumentException in Python.

Use UTF-8 to encode the name of column in Python 2, or it may failed to encode with default encoding ('ascii').

This PR also fix a bug when there is Java exception without error message.

Author: Davies Liu <davies@databricks.com>

Closes #7165 from davies/non_ascii and squashes the following commits:

02cb61a [Davies Liu] fix tests
3b09d31 [Davies Liu] add encoding in header
867754a [Davies Liu] support non-ascii character in column names

Capture the AnalysisException in SQL, hide the long java stack trace, only show the error message.

cc rxin

Author: Davies Liu <davies@databricks.com>

Closes #7135 from davies/ananylis and squashes the following commits:

dad7ae7 [Davies Liu] add comment
ec0c0e8 [Davies Liu] Update utils.py
cdd7edd [Davies Liu] add doc
7b044c2 [Davies Liu] fix python 3
f84d3bd [Davies Liu] capture SQL AnalysisException in Python API