Added Model Entry Points

[santacodes]

• Implemented Model Entry Points to create Model objects within the template and initialise them through entry points.
• Added the SPM model which can be initialised through the model entry points.
• A wrapper method to load a model object called models("modelname/authorname") is added.
  Example -
  To load the SPM model, after installing the pybamm_cookiecutter project, it can be accessed by calling, pybamm_cookiecutter.Model("SPM"). This would return an initialised model object of the SPM model.
• Added two basic tests for model entry points.

[valentinsulzer]

Nice to see this progressing! Good to prototype this in pybamm-cookiecutter but once it works well we should move the entry point logic to PyBaMM itself to match the ParameterValues entry points, for example users should be able to do pybamm.Model("author/name")

[santacodes]

Tweaked some things to unify both the entry points for parameter_sets and models, I noticed pybamm.parameter_sets is a singleton instance for loading parameter_sets, so I made some changes to make it a doubleton instance to accommodate Model Entry Points and Parameter Sets Entry Points in the same file, as in through the same entry point, just two different instances. I tested the logic and it seems to be working, if you create a third instance through the entry point it will just return you the object of the second instance(the most recent instance that was created).

You can load a Model by calling the pybamm_cookiecutter.Model('SPM') method, and load a parameter_set by doing pybamm_cookiecutter.parameter_sets["Chen2020"] the same way we do it in PyBaMM.

[agriyakhetarpal]

Slightly nitpicky, but since the API is being merged into one, could we test the model entry points and the parameter sets entry points in the same file?

[agriyakhetarpal]

I think this looks good to me and we have two approvals too, so I propose that we merge this so that we can start moving these files to the template. It's likely we'll revisit this code many times anyway, so once we are done, we can follow @valentinsulzer's suggestion in #23 (comment) to add the same things to the PyBaMM repository