Added REST API as a data source

[GeorgelPreput]

As per the recent Champions call, I've adapted my existing REST API data source to the PySpark Data Sources framework.

Currently supports as options:

• Sending headers
• No authentication, basic authentication, bearer token (via headers) and OAUTH2
• Both GET and POST
• Sending parameters
• Sending text data
• Sending JSON data

Output dataframe contains both the contents of the request, as well as the actual response, be that text data, JSON data or an error.

Known issues:

1. Having the protocol (http/https) in the call to .load("https://url-goes-here") makes Spark directly assume that the source should be Delta (at least in Databricks). Therefore, the URL passed must not have the protocol, and it can be added as an optional parameter: .option("protocol", "http"). The implicit default for this option is https.
2. Since both the request and the response get saved, a further # TODO might be to try to redact any tokens and client secrets from the output. Will add that upon request.

[allisonwang-db]

This is awesome! Thanks for adding it 👍

[GeorgelPreput]

Is there anything else I should add / modify?