SPARK-6063 MLlib doesn't pass mvn scalastyle check due to UTF chars in LDAModel.scala

Remove unicode characters from MLlib file.

Author: Michael Griffiths <msjgriffiths@gmail.com>
Author: Griffiths, Michael (NYC-RPM) <michael.griffiths@reprisemedia.com>

Closes #4815 from msjgriffiths/SPARK-6063 and squashes the following commits:

bcd7de1 [Griffiths, Michael (NYC-RPM)] Change \u201D quote marks around 'theta' to standard single apostrophe (\x27)
38eb535 [Michael Griffiths] Merge pull request #2 from apache/master
b08e865 [Michael Griffiths] Merge pull request #1 from apache/master

mllib/src/main/scala/org/apache/spark/mllib/clustering/LDAModel.scala

@@ -130,7 +130,7 @@ abstract class LDAModel private[clustering] {
 
   /* TODO
    * Compute the estimated topic distribution for each document.
-   * This is often called “theta” in the literature.
+   * This is often called 'theta' in the literature.
    *
    * @param documents  RDD of documents, which are term (word) count vectors paired with IDs.
    *                   The term count vectors are "bags of words" with a fixed-size vocabulary