Be cute with Option and getenv.

Creating these seems to take a while and clutters the output with Akka
stuff, so it would be nice to share them.

The previous version assumed that a CLASSPATH environment variable was
set by the "run" script when launching the process that starts the
ExecutorRunner, but unfortunately this is not true in tests. Instead, we
factor the classpath calculation into an extenral script and call that.

NOTE: This includes a Windows version but hasn't yet been tested there.

- Split SPARK_JAVA_OPTS into multiple command-line arguments if it
  contains spaces; this splitting follows quoting rules in bash
- Add the Scala JARs to the classpath if they're not in the CLASSPATH
  variable because the ExecutorRunner is launched with "scala" (this can
  happen when using local-cluster URLs in spark-shell)

Conflicts:
	run

Kafka fixes and DStream.count fix for master

Conflicts:
	.gitignore

Adding spark.streaming.blockInterval property

Surfacing decoders on KafkaInputDStream

Fixing count() in Spark Streaming

The old version reused the object within each task, leading to
overwriting of the object when a mutable type is used, which is expected
to be common in fold.

Conflicts:

	core/src/test/scala/spark/ShuffleSuite.scala

Shuffle fixes and cleanup

add Joblogger to Spark (on new Spark code)

Optimize PySpark worker invocation

- Exclude a version of ASM 3.x that comes from HBase
- Don't use a special ASF repo for HBase
- Update SLF4J version
- Add sbt-dependency-graph plugin so we can easily find dependency trees

fix typo and coding style in #638