Update documentation

[yhong123]

Resolved #131

Summary
This PR updates and reorganises the Datafaker documentation to improve clarity, navigation, and onboarding for users. In particular, it adds a new tutorial covering the workflow required by Paul and Shipra for generating synthetic Camino data from Parquet files.

Changes

• Added "Tutorial: Generate Synthetic Data from CSV and Parquet" (tutorial_parquet.rst), which serves as the primary guide for Paul and Shipra's use case.
  • The existing DuckDB documentation (duckdb.rst) has been retained for the time being and can be removed later if Tim agrees that tutorial_parquet.rst provides sufficient coverage.
• Restructured the documentation into two main sections: "Getting Started" and "Advanced".
• Updated index.rst to:
  • Refresh the introduction.
  • Remove outdated notes.
  • Revise the table of contents to reflect the new documentation structure.
• Renamed "Quick Start" to "Command-line Interface (CLI) Guide" to better represent the scope and content of quickstart.rst.
  • This change can be reverted if the original title is preferred.
• Renamed "Introductory Tutorial" to "Tutorial: Generate Synthetic Data from PostgreSQL" to improve clarity and maintain consistency with the naming convention used for other tutorials.
• Corrected a minor formatting issue in installation.rst.
• Removed "Example: Loan Data" (loan_data.rst) from index.rst because the example describes Turing's SqlSynthGen rather than Datafaker.
  • The file itself remains in the repository and can be removed later if Tim agrees.
• Added requirements.txt file for building the Sphinx documentation.
  • The file specifies the versions of Sphinx and the Mermaid extension required to build the documentation.
  • This simplifies the process of creating a virtual environment and installing the necessary dependencies before running poetry run make html in the docs directory.
• Updated /datafaker/README.md to include a link (https://safehr-data.github.io/datafaker/index.html) to the documentation GitHub Pages site.

Future Work
The "Advanced" section would benefit from additional review and reorganisation. Further documentation improvements can be addressed in a subsequent documentation-focused PR.

[tim-band]

This is good. It feels like it could be better to be clearer that 2 is automatic and that 3 is updating this. Something like:
2. Generate the default schema definition (orm.yaml file)
and leave 3 as it is perhaps?

[yhong123]

I have changed the line. See: https://github.com/SAFEHR-data/datafaker/pull/132/changes#diff-c421b880e222fce3eaee2200c40c91538814adb1a2bc0fd58e27abbb0f36e250R19

[tim-band]

orm.yaml is not updated by these operations.

[yhong123]

This line has been removed now.

[tim-band]

Perhaps add that this generates 100 lines in each table?

[yhong123]

I have added a new line. See: https://github.com/SAFEHR-data/datafaker/pull/132/changes#diff-c421b880e222fce3eaee2200c40c91538814adb1a2bc0fd58e27abbb0f36e250R244

[tim-band]

This is only for developers. Normal users will probably have installed DataFaker with pipx. I suppose "check it's installed"? I don't know.

[yhong123]

I have updated this line. See: https://github.com/SAFEHR-data/datafaker/pull/132/changes#diff-c421b880e222fce3eaee2200c40c91538814adb1a2bc0fd58e27abbb0f36e250R370

[tim-band]

Yes, this is good. Have a look at the suggestions and make improvements if you want. Thanks for doing this!

[yhong123]

Thank you Tim! I will make the changes based on your comments above and merge the code after you have merged the cross-table-interval branch.