(useful for jobs to pass their own JAR files to SparkContext).

Enable -optimize in the build

Fix code not to use code deprecated in Scala 2.9

Note that we use scala.Serializable introduced in Scala 2.9 instead of
java.io.Serializable. Also, case classes inherit from scala.Serializable by
default.

Workaround for scalac bug and publishTo configuration

Conflicts:
	project/build/SparkProject.scala

by specifying per-executor memory.

Until Spark is available in a Maven repository, this makes it easy to deploy to
e.g. GitHub pages by copying target/maven to the repository.

- Replaced modified Kryo version with the standard one augmented with
  the kryo-serializers package, which includes support for classes with
  no-arg constructors (that was why we had a modified Kryo before)
- The kryo-serializers version also fixes issue #72.
- Added a bunch of tests.
- Serialize maps and a few other common types properly by default.

Conflicts:
	core/src/main/scala/spark/Executor.scala
	core/src/main/scala/spark/RDD.scala

Conflicts:
	project/build.properties
	repl/src/main/scala/spark/repl/SparkInterpreterLoop.scala

Conflicts:
	project/build.properties