SPARK-821: Don't cache results when action run locally on driver

[SPARK-864]DAGScheduler Exception if we delete Worker and StandaloneExecutorBackend then add Worker

Caching the results of local actions (e.g., rdd.first()) causes the driver to
store entire partitions in its own memory, which may be highly constrained.
This patch simply makes the CacheManager avoid caching the result of all locally-run computations.

SPARK-884: Add unit test to validate Spark JSON output

Updating assembly README to reflect recent changes in the build.

This unit test simply validates that the outputs of
the JsonProtocol methods are syntactically valid JSON.

Link the Spark UI up to the Yarn UI 

Return the port the WebUI is bound to (useful if port 0 was used)

Conflicts:
	core/src/main/scala/org/apache/spark/ui/UIUtils.scala
	core/src/main/scala/org/apache/spark/ui/jobs/PoolTable.scala
	core/src/main/scala/org/apache/spark/ui/jobs/StageTable.scala
	docs/running-on-yarn.md

Conflicts:
	core/src/main/scala/org/apache/spark/deploy/master/Master.scala

Miscellaneous fixes for 0.8

- List higher-level projects that run on Spark
- Tweak CSS

Fix spark.io.compression.codec and change default codec to LZF

Allow PySpark to run on Windows