[SPARK-8763][PySpark] executing run-tests.py with Python 2.6 fails with absence of subprocess.check_output function

[cocoatomo]

Running run-tests.py with Python 2.6 cause following error:

Running PySpark tests. Output is in python//Users/tomohiko/.jenkins/jobs/pyspark_test/workspace/python/unit-tests.log
Will test against the following Python executables: ['python2.6', 'python3.4', 'pypy']
Will test the following Python modules: ['pyspark-core', 'pyspark-ml', 'pyspark-mllib', 'pyspark-sql', 'pyspark-streaming']
Traceback (most recent call last):
  File "./python/run-tests.py", line 196, in <module>
    main()
  File "./python/run-tests.py", line 159, in main
    python_implementation = subprocess.check_output(
AttributeError: 'module' object has no attribute 'check_output'
...

The cause of this error is using subprocess.check_output function, which exists since Python 2.7.
(ref. https://docs.python.org/2.7/library/subprocess.html#subprocess.check_output)

[AmplabJenkins]

Merged build triggered.

[AmplabJenkins]

Merged build started.

[SparkQA]

Test build #36280 has started for PR 7161 at commit cf4f901.

[SparkQA]

Test build #36280 has finished for PR 7161 at commit cf4f901.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds no public classes.

[AmplabJenkins]

Merged build finished. Test PASSed.

[davies]

You can use '_' for unused variable

[davies]

LGTM, merging this into master, thanks!