add inference_transpiler doc to develop branch

doc/fluid/api/transpiler.rst

@@ -14,6 +14,15 @@ DistributeTranspiler
     :members:
     :noindex:
 
+.. _api_fluid_transpiler_InferenceTranspiler:
+
+InferenceTranspiler
+-------------------
+
+..  autoclass:: paddle.fluid.transpiler.InferenceTranspiler
+    :members:
+    :noindex:
+
 .. _api_fluid_transpiler_memory_optimize:
 
 memory_optimize

python/paddle/fluid/transpiler/inference_transpiler.py

@@ -19,7 +19,7 @@ from ..framework import Program
 from ..executor import global_scope
 
 
-class InferenceTranspiler:
+class InferenceTranspiler(object):
     '''
     Convert the fluid program to optimized inference program.