Add tests for saving and loading every model

[huonw]

This PR test saving and loading of a tf.keras.Model instance for every model in StellarGraph.

It does so by adding a test_utils.model_save_load helper function that serialises and deserialisers a given model using various APIs provided by TensorFlow:

• saving:
  • tf.keras.Model.save
  • tf.keras.models.save_model
  • tf.saved_model.save
• loading: tf.keras.models.load_model (not tf.saved_model.load, because it doesn't create a Keras Model object)

Several of these new tests fail (they're included but marked as xfail), due to:

• Full batch models fail to save with "If specifying TensorSpec names for nested structures, either zero or all names have to be specified" #1251
• RGCN model fails to save with "'NoneType' object has no attribute 'replace'" #1252
• Cannot save an APPNP model: AttributeError: 'APPNPPropagationLayer' object has no attribute 'final_layer' #1680 (found in this PR)
• Cannot load a saved GCN_LSTM model: TypeError: __init__() missing 1 required positional argument: 'A' #1681 (found in this PR)

As those issues are fixed, the xfailing can be removed.

[codeclimate]

Code Climate has analyzed commit c1a5293 and detected 0 issues on this pull request.

View more on Code Climate.

[timpitman]

well done, glad that this is being tested even if there's xfails to address now.