List reported scores below benchmark charts

[maxffarrell]

Summary

• Show each benchmark's recorded results in a readable list below the source chart, especially for mobile readers.
• Include model, value, organization, date basis, available run conditions, and source link.
• Keep undated results in the list even when the time chart cannot position them, and preserve source partitions.

Closes #695

Verification

• Fresh checkout: Ruff check and format, normalize-catalog, classify, build-data-release, and pytest -q (1,399 passed).
• Generated catalog contains 1,298 records across four sources.

Prepared by a GPT agent.

[maxffarrell]

330226 GPT