the bool to tensor has a bug, if in init the training is False, the program not start.

IModel.Dictionary<string, float> evaluate(NDArray, NDArray, ...) is now IModel.Dictionary<string, float> evaluate(Tensor, Tensor, ...)
Merge Model.Evaluate.test_step_multi_inputs_function(...) and Model.Evaluate.test_function(...)

Note: An internal function need to add an explicit cast in Tensor

* Add check for dims of x and y in model.fit.

* Init the serialization of keras pb model.

* Add more facilities to the saved model framework.

* Add ListWrapper and ITrackable, and revise implmentations.

* Add serialized attributes.

* Implement layer serializations.

* Add lacked implementations (mainly MultiDeviceSaver).

* Support autograph.to_graph under graph mode.

* Add more implementations to the pb model save.

* Add more implementations to the keras part of pb model save.

* Refine some code after merge.

* Add two simple sequential test case of pb model save.

* Implement serializing attributes other keras arg definitions.

* Add alexnet pb save test.

* Check and refine the code.

---------

Co-authored-by: AsakusaRinne <AsakusaRinne@gmail.com>