Make the first example for the Sequential model pass

[cesarsouza]

Make the first example for a Sequential model given in https://keras.io/models/sequential/ pass as a unit test in the framework. The example is:

 model = Sequential()
 model.add(Dense(32, input_shape=(500,)))
 model.add(Dense(10, activation='softmax'))
 model.compile(optimizer='rmsprop',
       loss='categorical_crossentropy',
       metrics=['accuracy'])

and the unit test is located at https://github.com/cesarsouza/keras-sharp/blob/master/Tests/SequentialTest.cs#L28

[cesarsouza]

Almost there. The next step is to implement a C# equivelent of the _collect_metrics function and put it in Model.cs, line 577:

keras-sharp/Sources/Engine/Training/Model.cs

Line 577 in 222534b

private List<List<IMetric>> _collect_metrics(IMetric metrics, List<string> output_names)

[cesarsouza]

Done, the next goal now would be to implement a C# equivalent of the _masked_objective and put it in Model.cs, line 610:

keras-sharp/Sources/Engine/Training/Model.cs

Line 610 in 2c7f6a3

private IMetric _masked_objective(IMetric metric_fn)

[cesarsouza]

The execution path for the sequential model does not throw obvious exceptions anymore. It is now necessary to compare the complete state of the Sequential model against Python to determine properties that have not been set correctly.