Archive, part 1: an admin archives or wakes one repository from Settings and the API, and its page says "archived on <date>" with a Wake button #1838

Split from #1619. For: an owner can put a finished thing away without deleting it.

Done looks like:

• [ ] A named migration adds archived_at to repositories; the repository JSON carries archivedAt (null when not archived). This adds a field to the API and changes nothing else.
• [ ] PATCH /api/v1/repos/:owner/:name {archived: true|false} and Settings → Archive set and clear it, admins only. The form POST is same-origin checked.
• [ ] The repository page says "archived on <date>" with a Wake button for admins. Links, git clone and push keep working.
• [ ] No MCP change in this issue: the first try went red on test/mcp.test.ts (it speaks the protocol: initialize, a notification, ping, tools/list).
• [ ] Tests in a new file named in GATE: archive and wake round-trip; a non-admin gets a 403; a cross-origin POST is refused; the page shows the line. Run test/mcp.test.ts, test/apps.test.ts and bun x tsc --noEmit with it before pushing.

Parked by the crew: the check failed ("migrations are additive across a roll > the release record is newest first and never ahead of the schema") and the decider saw no point in another try (pull #1844 closed). Planner: split this smaller, or rewrite what done looks like; a person can also say what to try.

Parked by the crew: the check failed ("/boards > a panel keeps its chart and series: chosen in the form, saved in the file, drawn on the board") and the decider saw no point in another try (pull #1844 closed).

Parked twice since a person last spoke here, so it is the planner's now. Every reason:

1. the check failed ("migrations are additive across a roll > the release record is newest first and never ahead of the schema") and the decider saw no point in another try (pull #1844 closed)
2. the check failed ("/boards > a panel keeps its chart and series: chosen in the form, saved in the file, drawn on the board") and the decider saw no point in another try (pull #1844 closed)

The planner splits it, replaces it with a fresh issue, or defers it; a person can also say what to try. Nothing starts here until then, and the board lists it.

Replaced by #1998. Pull #1844 went red twice, both times only on main's own faults (the release record, fixed in 39e98bc9; the boards test, fixed in 43bf1e57). The fresh issue starts from c889df7b and names the stamped migration file.