Update create_datasets.ipynb to use Standard SQL and google-cloud-bigquery

[alixhami]

This PR makes progress toward #290

There are two updates in this PR:

• Change library for interacting with BigQuery to google-cloud-bigquery
• Change SQL syntax to Standard SQL (as opposed to Legacy SQL)

The hashing function changed in Standard SQL, which affected the results in this notebook. Take a look to see if any additional changes need to be made to the content given the change in results.

[lakshmanok]

When I tried it out recently, Datalab did not ship with a working version of google-cloud-bigquery (the make_exception package was missing) and so I had to pip install the package ...

See: https://github.com/GoogleCloudPlatform/training-data-analyst/blob/master/blogs/bqml/taxifare_bqml.ipynb

[alixhami]

@lakshmanok I'll follow up with them on the version of google-cloud-bigquery, but I'll add an installation to this PR as well.

Side note - in the notebook you linked, why is the google-cloud-bigquery library being installed but then called through pandas-gbq? I think it will be confusing for users if there is content with the client library (google-cloud-bigquery), Datalab's BQ module, and pandas-gbq all in the same repo.

[alexhanna]

This looks good and will most likely work for the the rest of the labs going forward.