If you fail to set the ElasticNet parameter to 0.0, you will get an error because L2 regularization is used when solving through normal equations in Spark (as of this writing).

Documentation for Spark 2.0 related to isotonic regression can be found at: http://spark.apache.org/docs/latest/api/scala/index.html#org.apache.spark.ml.regression.LinearRegression and http://spark.apache.org/docs/latest/api/scala/index.html#org.apache.spark.ml.regression.LinearRegressionModel

The model summary can be found at: http://spark.apache.org/docs/latest/api/scala/index.html#org.apache.spark.ml.regression.LinearRegressionSummary