[SPARK-8765] [MLLIB] Fix PySpark PowerIterationClustering test issue

PySpark PowerIterationClustering test failure due to bad demo data.
If the data is small,  PowerIterationClustering will behavior indeterministic.

Author: Yanbo Liang <ybliang8@gmail.com>

Closes #7177 from yanboliang/spark-8765 and squashes the following commits:

392ae54 [Yanbo Liang] fix model.assignments output
5ec3f1e [Yanbo Liang] fix PySpark PowerIterationClustering test issue

python/pyspark/mllib/clustering.py

@@ -282,18 +282,30 @@ class PowerIterationClusteringModel(JavaModelWrapper, JavaSaveable, JavaLoader):
 
     Model produced by [[PowerIterationClustering]].
 
-    >>> data = [(0, 1, 1.0), (0, 2, 1.0), (1, 3, 1.0), (2, 3, 1.0),
-    ...     (0, 3, 1.0), (1, 2, 1.0), (0, 4, 0.1)]
+    >>> data = [(0, 1, 1.0), (0, 2, 1.0), (0, 3, 1.0), (1, 2, 1.0), (1, 3, 1.0),
+    ... (2, 3, 1.0), (3, 4, 0.1), (4, 5, 1.0), (4, 15, 1.0), (5, 6, 1.0),
+    ... (6, 7, 1.0), (7, 8, 1.0), (8, 9, 1.0), (9, 10, 1.0), (10, 11, 1.0),
+    ... (11, 12, 1.0), (12, 13, 1.0), (13, 14, 1.0), (14, 15, 1.0)]
     >>> rdd = sc.parallelize(data, 2)
     >>> model = PowerIterationClustering.train(rdd, 2, 100)
     >>> model.k
     2
+    >>> result = sorted(model.assignments().collect(), key=lambda x: x.id)
+    >>> result[0].cluster == result[1].cluster == result[2].cluster == result[3].cluster
+    True
+    >>> result[4].cluster == result[5].cluster == result[6].cluster == result[7].cluster
+    True
     >>> import os, tempfile
     >>> path = tempfile.mkdtemp()
     >>> model.save(sc, path)
     >>> sameModel = PowerIterationClusteringModel.load(sc, path)
     >>> sameModel.k
     2
+    >>> result = sorted(model.assignments().collect(), key=lambda x: x.id)
+    >>> result[0].cluster == result[1].cluster == result[2].cluster == result[3].cluster
+    True
+    >>> result[4].cluster == result[5].cluster == result[6].cluster == result[7].cluster
+    True
     >>> from shutil import rmtree
     >>> try:
     ...     rmtree(path)