- Fixed an compilation error due to wrong 'import' of legacy lzf libraries in DfsBroadcast.scala

 - Updated to use ning libraries.
 - Passes all unit tests

src/scala/spark/Broadcast.scala

@@ -14,8 +14,6 @@ import scala.collection.mutable.Map
 import org.apache.hadoop.conf.Configuration
 import org.apache.hadoop.fs.{FileSystem, Path, RawLocalFileSystem}
 
-import com.ning.compress.lzf.{LZFInputStream, LZFOutputStream}
-
 @serializable
 trait Broadcast[T] {
   val uuid = UUID.randomUUID

src/scala/spark/DfsBroadcast.scala

@@ -7,7 +7,7 @@ import java.util.UUID
 import org.apache.hadoop.conf.Configuration
 import org.apache.hadoop.fs.{FileSystem, Path, RawLocalFileSystem}
 
-import spark.compress.lzf.{LZFInputStream, LZFOutputStream}
+import com.ning.compress.lzf.{LZFInputStream, LZFOutputStream}
 
 @serializable 
 class DfsBroadcast[T](@transient var value_ : T, isLocal: Boolean)