[Issue 894] analytics documentation

[widal001]

Summary

Reorganizes and expands the documentation to reflect the structure of docs for API and Frontend sections of the codebase.

Fixes #894

Time to review: 15 mins

Changes proposed

What was added, updated, or removed in this PR.

• Adds the following docs to documentation/analytics/
  • development.md How to install and run the analytics package locally and complete common tasks, like:
    • Adding a new dataset
    • Adding a new metric
  • formatting-and-linting.md Describes the tools used for linting and formatting the code
  • testing.md Describes how to run and write tests for the test suite
  • usage.md Describes how to run the report and view the results
  • technical-overview.md Describes the key technologies used in the project
  • metrics/Section with more detailed explanations of how metrics are calculated, including expected inputs and outputs
• Reorganizes docs in analytics/README.md

Context for reviewers

Testing instructions, background context, more in-depth details of the implementation, and anything else you'd like to call out or ask reviewers. Explain how the changes were verified.

I tried to mirror the structure of the docs for the other sections of the codebase:

• api/ and documentation/api/
• frontend/ and documentation/frontend/

Additional information

Screenshots, GIF demos, code examples or output to help show the changes working as expected.

[AlexanderStephensonUSDS]

I like the inclusion of example data to walk through the metrics.

As we think about future metrics to the project, is coupling steps with examples the approach we want to take for all metrics?

Some reasons for the current structure as I see it:

• an individual looking at the metric gets a clear understanding of the example calculation steps required
• an individual sees how the inputs connect to the outputs.
• Explanation of step happens with an example.

Some reasons against the current structure:

• Readability right now requires scrolling through large amount of text that might not be essential for understanding. The percent-complete metric right now is the same metric to two different examples.
• The steps lack a table of contents, also affecting readability.

An intermediate compromise might be to have a section that is the brief "how" section of the metric calculation itself (e.g. mean = sum(x)/len(x)) and then the examples are split into calculation steps.

[lucasmbrown-usds]

Extremely thorough and comprehensive! Kudos!