Automate TigerTag JSON database updates

[TigerTag-Project]

Keeping the TigerTag tables in src/tag/tigertag/database/ up to date currently means running download_json.py by hand and opening a PR (as in #17 and #31). This PR automates it.

download_json.py

It now follows the sync strategy described in the TigerTag RFID Guide (section 2.0):

• Fetches the last_update manifest first, and only downloads the tables whose timestamp changed. The manifest is stored as database/last_update.json.
• Uses the TigerTag API, and falls back to the GitHub mirror of the guide (synced every 6 hours) if the API is unreachable.
• Checks each download is a non-empty JSON array before replacing the file, so a failed run never leaves a broken table behind.
• Writes the tables indented, one entry per line. Today each table is a single line, so every update shows as "1 line changed" and is impossible to review.
• New options: --force re-downloads every table, and --summary FILE writes a markdown list of added, removed and changed entries.

The parser is unchanged: registry.py reads the same files.

Workflow .github/workflows/update-tigertag-json.yml

• Runs every Monday at 06:00 UTC, and on demand (workflow_dispatch).
• Syncs the tables. If something changed, it runs pytest, then opens (or updates) a pull request from the automation/update-tigertag-json branch, with the change summary as the description.
• Tests run inside this workflow because pull requests opened with GITHUB_TOKEN do not trigger the pytest workflow.

One setting is needed for the workflow to open pull requests: Settings → Actions → General → Workflow permissions → tick "Allow GitHub Actions to create and approve pull requests".

Notes

• The first run downloads every table, since no last_update.json exists yet, and reformats them. Its PR will therefore also include whatever Update TigerTag JSON database #31 has not merged yet.
• Tested locally: first run, a second run with nothing to download, and the GitHub fallback with the API unreachable. Tests: 95 passed, 1 skipped.

[TigerTag-Project]

Hi @suchmememanyskill, we're the TigerTag team. This PR automates the TigerTag JSON database updates, so future updates no longer need manual PRs like #17.

We tested the workflow end-to-end on our fork before submitting it: it synced all 7 tables from the TigerTag API, ran pytest (all green), and opened the update PR by itself, with a readable summary of added and changed brands and materials. See the test run and the PR it opened.

To enable it after merging, one setting is needed: Settings → Actions → General → Workflow permissions → "Allow GitHub Actions to create and approve pull requests". The workflow then runs every Monday at 06:00 UTC, and can also be started by hand from the Actions tab.

Two related PRs are also open:

• Update TigerTag JSON database #31 refreshes the JSON database now (+20 brands, +17 materials). Please merge it first: it brings the tables up to date right away, whatever you decide about this automation. The two PRs don't conflict, since this one doesn't touch the JSON files.
• TigerTag: custom message and bed temperature (supersedes #18) #32 takes over Add GenericFilament fields and update TigerTag reading #18 by @physicsG (custom message and bed temperature), with the fixes it needed to pass the tests on the current main.

Happy to adjust anything!