Fix AbstractQuerySet.order_by to consider existing conditions

Scylla does not tolerate queries with repeated column, like one below:
`ORDER BY "attempt_id" ASC, "attempt_id" DESC`

Author: dkropachev

cassandra/cqlengine/query.py

@@ -819,12 +819,19 @@ class Comment(Model):
             clone._order = []
             return clone
 
-        conditions = []
-        for colname in colnames:
-            conditions.append('"{0}" {1}'.format(*self._get_ordering_condition(colname)))
-
         clone = copy.deepcopy(self)
-        clone._order.extend(conditions)
+
+        for col_name in colnames:
+            col_name, order = self._get_ordering_condition(col_name)
+            col_name = '"{0}"'.format(col_name)
+            # Check if there is existing condition that targets same column and replace it if it's order is different
+            for cond_id, existing_condition in enumerate(clone._order):
+                if col_name in existing_condition:
+                    if not existing_condition.endswith(order):
+                        clone._order[cond_id] = '{0} {1}'.format(col_name, order)
+                    break
+            else:
+                clone._order.append('{0} {1}'.format(col_name, order))
         return clone
 
     def count(self):