test: clean up the paid eval lane (B1-B4, B6, B7)

- B1: delete paid files that assert nothing or cannot pass meaningfully:
  skill-llm-eval-spec and skill-e2e-spec-execute (test.todo), gemini-e2e
  (+ gemini-session-runner; no gemini CLI in CI), ship-idempotency (red
  since v1.63), the two opus-4-7 *-sonnet overlay wrappers, conductor-prose
  (+ its source-evaluation replay), codex-e2e-plan-format; drop their keys,
  scripts and census rows.
- B2: skill-llm-eval grades browse/sections/command-list.md with one union
  judge that also carries the baseline score pin; regression-vs-baseline
  deleted (paid run: pass, c4/c4/a4).
- B3: memory-pipeline, ios-qa, ios-qa-swift-build and plan-tune-cathedral
  make no model calls; renamed out of the paid glob so they run on every
  PR. Swift builds need GSTACK_TEST_SWIFT=1; device stub deleted.
- B4: codex-e2e*, outside-voice, aside and ios-device cannot run in the CI
  image; excluded from the weekly lane with a tracked re-entry condition.
- B6: fold opus-47's negative routing controls into skill-routing-e2e
  journey-negatives (paid run: 3/3 unrouted) and delete the file.
- B7: delete the never-green brain-privacy-gate eval; a free
  gstack-skill-start test now proves consent precedes artifacts egress.

test/fixtures/conductor-prose-ao.json

@@ -1,24 +0,0 @@
-{
-  "publicDecisionTail": "D1—ReducethepricingtiertoStripe-nativepieces,orbuildallfourcomponents?\r\rReplywithaletter(Conductorsession,sothisbriefisinprose).Project/branch:gstackedinburgh-v1,reviewingan\rexternal pricing-tier plan.\r\rELI10:Youwantmoredeveloperstosignup,soyou'readdingacheaperplan.Thedraftbuildsfournewthingstodo\rit: a Stripe price, a custm pricing web age,r own atabse tbletht remmbers who paid fowhat, ad a fast\rn-memory copy of tht table. Strpealready ships the table (Entitlements)and th page (rcing table). Every exra\rcopy of \"wh is allowed to dowhat\" is another place that can disagre withStripe when a card fails or awebhook\rarriveslate. The stakes: a developer who paidand sees \"upgrae required,\"or ono cnelled andkeeps access.\r\rStakesifwepickwrong:threesourcesoftruthforentitlementswithnoinvalidationdesignmeanssilentaccessbugs\rtht onlyshow upas angry support tickets.\r\rRecommendation:Abecauseitkeepseveryuser-visibleoutcomeofthedraftwhiledeletingthetwocomponentsmost\rlikely to causeentitlemnt drift,and it fits \"engineered enough\" an \"smallestiff that cleanly exressethe\rchange.\"\r\rCompleteness:A=9/10,B=7/10.Cisahold,notabuildoption,sonoscore.\r\rA)Stripe-nativeslimbuild(recommended).Completeness9/10.Human:~1week/CC:~1to2hours.AddoneStripePrice\rplus a Fature-to-Product mapping, mirror active_entitlement_sumryinto a singleexisting cstomer table hrough an\ridempotent webhok, gate fetures off that olumn, ship the pricing surfacebehinda feature flag (Stripe pricin\u001b[?25h",
-  "actualFlags": {
-    "proseAUQEverObserved": false,
-    "waitingEverObserved": true,
-    "scopeGateAutoSelectObserved": true
-  },
-  "owner": {
-    "sessionId": "4e820ffc-b6dd-47d7-b074-5c4b0a3cbb68",
-    "commandStartedAt": 1789041170391,
-    "nativePolledAt": 1789041383968,
-    "captureAt": "2026-09-10T11:56:33.483Z"
-  },
-  "provenance": {
-    "observation": ".context/ship-source-ao-delta-paid-20260910-v1/conductor-first-failure-ledger-v1/observation.json",
-    "observationSha256": "da8c8a342f44ae06bfdd092f1bdb6a395765553d65ac99eecaaa35dedb34545d",
-    "visible": ".context/ship-source-ao-delta-paid-20260910-v1/conductor-first-failure-ledger-v1/terminal.visible.log",
-    "visibleSha256": "2a2fca3e8d1b51694d76a3dbcfac524626f856e4ebf38f4bd093346edb64bf82",
-    "startCharacterOffset": 78469,
-    "publicTailSha256": "e551890b1d839cee7d7dce05120b8dc05a54f27fe8f69617aa99384c49316604",
-    "limit": "Exact public partial D1 tail; no completed native prose message retained. A later second option in tests is synthetic.",
-    "offsetUnit": "Unicode code points in decoded retained bytes, without newline normalization"
-  }
-}

test/fixtures/eval-baselines.json

@@ -1,6 +1,4 @@
 {
-  "command_reference": { "clarity": 4, "completeness": 3, "actionability": 4 },
-  "snapshot_flags": { "clarity": 4, "completeness": 4, "actionability": 4 },
   "browse_skill": { "clarity": 4, "completeness": 4, "actionability": 4 },
   "qa_workflow": { "clarity": 4, "completeness": 4, "actionability": 4 },
   "qa_health_rubric": { "clarity": 4, "completeness": 3, "actionability": 4 }

test/fixtures/overlay-nudges.ts

@@ -305,50 +305,6 @@ export const OVERLAY_FIXTURES: OverlayFixture[] = [
     pass: lowerIsBetter20Pct,
   },
 
-  {
-    id: 'opus-4-7-effort-match-trivial-sonnet',
-    overlayPath: 'model-overlays/opus-4-7.md',
-    model: 'claude-sonnet-4-6',
-    trials: 10,
-    concurrency: 3,
-    direction: 'lower_is_better',
-    maxTurns: 8,
-    setupWorkspace: (dir) => {
-      fs.writeFileSync(
-        path.join(dir, 'config.json'),
-        '{"name": "demo", "version": "1.0.0"}\n',
-      );
-    },
-    userPrompt: "What's the version in config.json? Return only a JSON object with the version key and its exact string value. " +
-      "The final message is consumed directly by JSON.parse: return the exact JSON object, with no Markdown fences and no other prose.",
-    metric: reportedThinkingTokens,
-    metricName: 'reported_thinking_tokens',
-    verify: (r) => assertFinalJson(r, { version: '1.0.0' }),
-    comparison: { direction: 'lower_is_better', minimum: 0 },
-    pass: lowerIsBetter20Pct,
-  },
-
-  {
-    id: 'opus-4-7-literal-interpretation-sonnet',
-    overlayPath: 'model-overlays/opus-4-7.md',
-    model: 'claude-sonnet-4-6',
-    trials: 10,
-    concurrency: 3,
-    direction: 'higher_is_better',
-    allowedTools: ['Read', 'Glob', 'Grep', 'Bash', 'Edit', 'Write'],
-    maxTurns: 15,
-    setupWorkspace: setupLiteralWorkspace,
-    userPrompt: 'Fix the failing tests. Preserve the specified behavior and repair the implementation.',
-    metric: (_r, dir, deadlineAt) => {
-      if (!dir) throw new Error('literal fixture metric needs its workspace');
-      return correctLiteralTargets(dir, deadlineAt);
-    },
-    metricName: 'correct_target_behaviors',
-    taskCorrect: (metric) => metric === 3,
-    allowedChanges: ['src/auth.ts', 'src/billing.ts', 'src/notifications.ts'],
-    comparison: { direction: 'higher_is_better', minimum: 0, maximum: 3 },
-    pass: higherIsBetter20Pct,
-  },
 ];
 
 // Validate at module load so a broken fixture fails fast at test startup,