Notify standalone deploy client of application death.

Conflicts:
	core/src/main/scala/spark/storage/BlockManagerMaster.scala
	core/src/test/scala/spark/scheduler/DAGSchedulerSuite.scala

Usually, this isn't necessary since the application will be removed
as a result of the deploy client disconnecting, but occassionally, the
standalone deploy master removes an application otherwise.

Also mark applications as FAILED instead of FINISHED when they are
killed as a result of their executors failing too many times.

Fixed SPARK-706: Failures in block manager put leads to read task hanging.

Track assorted metrics for each task, report summaries to user at stage completion

Worker address was getting removed when removing an app.

Avoid bind failure in InputStreamsSuite

Instead of failing to bind to a fixed, already-in-use port, let the OS choose an available port for TestServer.

bump version to 0.7.1-SNAPSHOT in the subproject poms

hanging.

Conflicts:
	core/src/test/scala/spark/scheduler/DAGSchedulerSuite.scala

createNewSparkContext should use sparkHome/jars/environment.

This fixes a bug introduced by Matei's recent change.

Very minor change in a testcase

Fixed bug in BlockManager and added a testcase

For some reason, ScalaDoc complained about no such constructor for
StreamingContext; it doesn't seem like an actual Scala error but it
prevented sbt publish and from working because docs weren't built.