[Task]: Task 3 — Improve Quail ingest error visibility (surface bounce reasons / ingest failures in admin UI)

[devils-shadow]

Problem

Quail UI showed “nothing” while Postfix was actively bouncing due to ingest script failure. That wastes time and causes humans to invent theories. Humans already do that unprompted.

Goal

Make ingest failures obvious from within Quail.

Proposed change

• Add an “Ingest health” panel in admin:
  • last N ingest attempts, success/failure, and error summary.
• Option A (minimal): Quail ingest script writes structured logs to a known file or journal tag; Quail reads last lines.
• Option B: ingest writes to DB table (preferred if Quail already has DB for messages).

Acceptance criteria

• If ingest fails (simulate by breaking the venv path), admin UI shows recent failure with the error cause.
• When fixed, admin UI shows successful ingests.

Test plan

• Force failure: temporarily point ingest to non-existent python path.
• Confirm UI shows error.
• Restore, confirm UI shows success.