karvey-testlisted

# Karvey Test ## Purpose Run the full post-implementation test plan: unit tests per layer and E2E tests of the complete flow. **The contract is §6 (test coverage plan) of `architecture.md`**: every case it lists is run or reported as not run, with the reason. Document evidence in `docs/test_evidence.md`. ## Execution steps ### Step 1 — Load context Enter the phase: `python3 "${CLAUDE_PLUGIN_ROOT}/scripts/karvey-state.py" advance "{change-id}" test` (refused while an earlier gate is open; on a resumed run `next` shows it is already in `test`). Read: - `docs/spec/changes/{change-id}/architecture.md` — §6 is the test contract - `docs/spec/changes/{change-id}/requirements.md` - `docs/spec/changes/{change-id}/mockup/` if the change has one (to map E2E flows) - `docs/spec/changes/{change-id}/tasks.md` Also read `docs/spec/project.json` and obtain the `targets` field (see `../karvey/rules/targets.md`). The actual runtime in which the E2E tests run depends on the declared target: browser (web), simulator/device (iOS/Android), terminal (CLI), HTTP client (API), hardware/emulator (embedded). **Do not assume "web" by default** — a project may have multiple targets. Detect stack: `package.json`, `requirements.txt`, `pyproject.toml`, `go.mod`, `pom.xml`, `Gemfile`, `Cargo.toml`, or other project configuration files. Identify: - Backend language and framework - Database type and access pattern (ORM, SPs, direct queries) - Available unit test framework - Available E2E framework - AP