Add get_by_id to EdFiEndpoint

[ea-mtenhoor]

This PR adds a get_by_id method to EdFiEndpoint, allowing the retrieval of a single resource record by its Ed-Fi ID. A similar result is already possible by passing a dictionary with id to the params of get_rows. There is no practical performance improvement with either method, but there are some differences.

• GET with params ends up making two calls to the Ed-Fi API.
  /data/v3/ed-fi/schools?id=44cbdc9355c74afe90262493a2aba769 and
  /data/v3/ed-fi/schools?id=44cbdc9355c74afe90262493a2aba769&pageToken=LTIxNDc0ODM2MzgsMjE0NzQ4MzY0Nw&pageSize=100
• GET with params generates a SQL statement with unnecessary where clauses, including limit and offset.
• GET with params returns a list with (probably) one object
• GET with params is more code on the implementer side
• GET by ID makes a single call.
  /data/v3/ed-fi/schools/44cbdc9355c74afe90262493a2aba769
• GET by ID only has one SQL query parameter, the ID.
• GET by ID returns a single dictionary.
• GET by ID is less code on the implementer side

schools = edfi_api.resource('schools')
get_params = list(schools.get_rows(params={'id': '44cbdc9355c74afe90262493a2aba769'}))[0]
get_by_id  = schools.get_by_id('44cbdc9355c74afe90262493a2aba769')

[jayckaiser]

Are there any params that are relevant to include when querying by ID directly? If not, I would want to omit these from being used to avoid the possibility of someone passing mutually-exclusive arguments and parameters (e.g., different IDs, etc.).

[jayckaiser]

To confirm, if the ID passed is undefined, will this return an empty dictionary?