Community page with CSV/JSON download

[paccloud]

What to build

Anyone, signed in or not, can browse the community dataset (all approved yields) on the Community page, read from the public collection. They can also download it as CSV or JSON. Each yield is attributed only by display name or "Anonymous". The page and both files carry the CC BY 4.0 notice (ADR 0004 in #122).

Values in the CSV are user-supplied, so any cell starting with =, +, - or @ is neutralized and a spreadsheet will not run it as a formula. This matches today's server export.

A stable download address for other tools comes later with versioned releases; it is not part of this slice.

Acceptance criteria

• The Community page lists approved yields without sign-in, and can be filtered by species
• Only public fields are shown: species, conversion, yield and display name (or "Anonymous")
• CSV and JSON downloads contain the same fields
• CSV cells starting with =, +, - or @ are neutralized, with the current regression test ported
• The page shows the CC BY 4.0 notice with an attribution line
• Each downloaded file includes the notice: a header comment in the CSV, a license field in the JSON
• No other identifying field, such as organization or the free-text source note, appears anywhere
• The page works offline with the last-loaded data

Blocked by

• Submit yields for review, and the Review page #128