Investigate implementing frictionlessdata.io : table-schema for resources.

More detail [here](https://github.com/bcgov/ckanext-bcgov/wiki/Roadmap#integration-of-friction-less-data)

We would like users to provide a schemas for resources uploaded to the BCDC.
This schema would follow the [table-schema spec](http://specs.frictionlessdata.io/table-schema/) proposed on [FrictionlessData.io](https://frictionlessdata.io)
They should be able to provide their own schema in `json` or have a tool predict what the schema is - similar to what messy tables does when a `csv` is pushed to the data store.

http://specs.frictionlessdata.io/table-schema/

Read the linked BCDC roadmap entry and the FrictionlessData table-schema specification first. Inspect how this CKAN extension handles uploaded resources and how messy tables predicts schemas for CSV data. Done means users can supply a table schema in JSON or receive a predicted schema for an uploaded resource, with the integration behavior defined and tested.

Written by the indexing model from the issue text.