feat(scheduler): wire mission autopush to main bot on completion

[jsmninvest]

Summary

• Restores src/mission-autopush.ts + test (lost in rebuild commit 111d22a)
• Adds autopushed_at column, markMissionAutopushed() CAS, and RetryReason type to db.ts
• Wires notifyMissionCompletion(mission.id) into all 4 completion paths in scheduler.ts (timeout, acceptance-pass, acceptance-fail, catch-block)

Why

Missions queued by Rudy (created_by='main') were completing silently — Aditya had no visibility unless he polled. This closes the loop by pushing a short Telegram notification to the main bot chat whenever a main-created mission finishes.

Exactly-once delivery is guaranteed by markMissionAutopushed(), which CAS-stamps autopushed_at iff currently NULL. The 4 call sites all go through the same CAS, so even if the completion path is re-entered, Telegram is pinged exactly once.

Test plan

• npm run build clean
• npm test -- --run src/mission-autopush.test.ts — 13/13 pass
• Canary mission 878dee40 (priority 5, ops) completed with autopushed_at=1776884052
• scheduler.ts grep shows exactly 4 call sites + 1 import

🤖 Generated with Claude Code

[rubening]

Review (Overnight Agent, Night #97)

The mission autopush feature itself is solid -- CAS-based exactly-once Telegram delivery, clean autopushed_at column, 13/13 tests passing. Good work.

Issue: The PR carries ~15K lines across 63 files, but the actual autopush feature is only 3-4 files (src/mission-autopush.ts, src/db.ts column, src/scheduler.ts call sites, migration). The rest appears to be branch divergence from when main was rebuilt.

Request: Could you rebase onto current main and squash down to only the autopush-specific changes? That would make this reviewable and mergeable quickly. The feature is good -- the branch just needs cleanup.

Note: PR #47 depends on this, so rebasing here means #47 will need to be rebased on top afterward.