bug-workflowlisted

# bug-workflow Create one durable Bug Case from test evidence, then classify before choosing an owner. This is a triage workflow: it does not diagnose every failure, change product code, or run the repaired artifact. Write it from `bug-case-template.md` to `.harness-ship/bugs/<BUG-ID>.md`. ## 1 — Intake one Bug Case Start from the failing attempt in `.harness-ship/candidates/<short-sha>/ledger.md` — that directory takes the contract's **Candidate identifier** where its **Release surface owner** is not this repository — and the validated handoff beside it at `handoff.md`. Preserve: - Test Contract revision, scenario ID, and originating work; - full source SHA — or the candidate identifier where the contract's **Release surface owner** is not this repository — exact tested artifact/environment revision, and provenance receipt; - expected and actual behaviour, reproducibility, and durable evidence. Assign one **stable BUG-ID** once. If the tracker publishes the case, use its ID; otherwise assign a portable local BUG-ID. Every later classification, handoff, diagnosis, and retest is an append-only event under that same ID. Never replace the Bug Case with a new defect and never rewrite history. Keep these independent fields: - **phase** — where triage currently is; - **classification** — what kind of finding the evidence supports; and - **disposition** — the current routing decision. ## 2 — Classify before diagnosis Choose exactly one classification and append the eviden