[SPARK-21241][MLlib]- Add setIntercept to StreamingLinearRegressionWithSGD in Pyspark.

StreamingLinearRegressionWithSGD class in PySpark is missing the setIntercept Method which offers the possibility
to turn on/off the intercept value. API parity is not respected between Python and Scala. We add the setIntercept
Method to StreamingLinearRegressionWithSGD class which calls setIntercept Method in LinearRegressionModel class
in order to turn on/off the intercept. A big thanks to Matthieu CANEILL for his precious help in solving the issue.
This patch was tested by running all tests with ./dev/run-tests and by manual tests.

Author: SoulGuedria

python/pyspark/mllib/regression.py

@@ -199,6 +199,10 @@ def load(cls, sc, path):
         model = LinearRegressionModel(weights, intercept)
         return model
 
+    @since("2.3.0")
+    def setIntercept(self, intercept):
+        self._intercept = intercept
+
 
 # train_func should take two parameters, namely data and initial_weights, and
 # return the result of a call to the appropriate JVM stub.
@@ -795,6 +799,11 @@ def __init__(self, stepSize=0.1, numIterations=50, miniBatchFraction=1.0, conver
         super(StreamingLinearRegressionWithSGD, self).__init__(
             model=self._model)
 
+    @since("2.3.0")
+    def setIntercept(self, intercept):
+        """Set if the algorithm should add an intercept"""
+        self._model.setIntercept(intercept)
+
     @since("1.5.0")
     def setInitialWeights(self, initialWeights):
         """