v1.89.0.0 feat: add shared-code extraction audit (#2925)

* feat: bind shared-code review advice to source and branch

* feat: add shared-code extraction audit and scoped review checks

* test: recognize complete source reads and explicit coverage legends

* chore: bump version and changelog (v1.88.0.0)

Co-Authored-By: OpenAI Codex <noreply@openai.com>

* test: capture native review questions and retain public evidence

Capture the actual first public native question with strict ownership and display matching. Preserve terminal failures and raw evidence, and retain SDK completion checks.

* test: recognize verified review evidence and complete fixtures

Recognize complete source and diagram evidence, concrete design and developer-experience decisions, and the complete planted scenario contracts. Preserve negative controls and grading thresholds.

* fix: preserve decision brief structure in native questions

Keep the required pros-and-cons heading and final Net field in native question text. Regenerate host outputs and document the release and evaluation repairs.

Co-Authored-By: OpenAI Codex <noreply@openai.com>

* docs: update project documentation for v1.88.0.0

Co-Authored-By: OpenAI Codex <noreply@openai.com>

* fix: correct eval retry accounting and ship workflow gates

* fix: capture native eval evidence and stabilize CI fixtures

* fix: keep shared-code eval skips read-only

Choose explicit no-change answers instead of mixed fix/preservation options.
Reuse the bounded revalidation prompt for path fixtures so required review
metadata is available without repeated discovery. Preserve source checks,
retry limits, and failed native terminal outcomes.

Add captured-question and callback regressions, plus evaluation selection
coverage for the affected fixtures.

---------

Co-authored-by: OpenAI Codex <noreply@openai.com>

test/fixtures/design-count-sep21-header-first-call.json

@@ -0,0 +1,41 @@
+{
+  "provenance": {
+    "source": "September 21 design-count facts diagnostic",
+    "runId": "shared-libs-design-facts-sep21",
+    "sourceCallSha256": "37ff256f4b3485c59bf9e823bfa1b4d2c79e39757fdb68397339041b416126c2"
+  },
+  "calls": [
+    {
+      "sessionId": "f46fbe04-3291-4d12-98d7-bddda43b1681",
+      "toolUseId": "toolu_015hpK1GGLDc2UpK1e8RRodu",
+      "questions": [
+        {
+          "header": "Issue 1: Save",
+          "question": "D1 — Should Save become the only filled primary action in the header, per DESIGN.md?\n\nProject/branch/task: gstack-plan-count on main, account-settings form conformance update.\nELI10: Right now Save, Reset, Cancel and Export look identical. A user scanning the header cannot tell which one finishes their task, and Reset and Cancel are destructive-ish choices sitting at the same visual weight as Save. Users satisfice: they click the first plausible button. Equal weight makes a wrong click as likely as a right one.\nStakes if we pick wrong: accidental Reset/Cancel taps on mobile, and a “which button do I press?” pause on every visit that drains goodwill.\nRecommendation: 1A because DESIGN.md already names the exact tokens and it keeps one accent color on the page (Hierarchy as service; Use conventions).\nCompleteness: 1A=10/10, 1B=8/10, 1C=2/10\nNet: one filled button versus a second tier that DESIGN.md does not define, versus leaving the page with no primary action.",
+          "multiSelect": false,
+          "options": [
+            {
+              "label": "1A) DESIGN.md primary (recommended)",
+              "description": "Save: filled #1d4ed8 with white text, 44px. Reset, Cancel, Export: neutral ghost buttons. ✅ Uses the exact approved tokens; nothing new to design or maintain. ✅ Single accent color keeps the header calm and the primary action obvious in a 3-second scan. ❌ Export loses any visual distinction from the two discard actions; its label must carry the meaning alone. (human: ~1h / CC: ~5min)"
+            },
+            {
+              "label": "1B) Three tiers: filled, outlined, ghost",
+              "description": "Save filled #1d4ed8; Export outlined #1d4ed8 border; Reset and Cancel ghost. ✅ Separates the non-destructive Export from the two discard actions at a glance. ✅ Still leaves Save as the only filled button. ❌ Adds a button variant DESIGN.md does not define, so it needs a new token and a DESIGN.md amendment. ❌ Two blue elements in one row compete for the first glance. (human: ~3h / CC: ~15min)"
+            },
+            {
+              "label": "1C) Keep four equal buttons",
+              "description": "No change; rely on position (Save first) alone. ✅ Zero implementation cost. ✅ No risk of visual drift from the current build. ❌ Leaves a known DESIGN.md violation and no primary action; on 640px and below, Save is full-width but still looks like the others. ❌ Lowers Pass 1 and Pass 5 final scores."
+            }
+          ]
+        }
+      ],
+      "answered": true,
+      "failed": false,
+      "answers": {
+        "D1 — Should Save become the only filled primary action in the header, per DESIGN.md?\n\nProject/branch/task: gstack-plan-count on main, account-settings form conformance update.\nELI10: Right now Save, Reset, Cancel and Export look identical. A user scanning the header cannot tell which one finishes their task, and Reset and Cancel are destructive-ish choices sitting at the same visual weight as Save. Users satisfice: they click the first plausible button. Equal weight makes a wrong click as likely as a right one.\nStakes if we pick wrong: accidental Reset/Cancel taps on mobile, and a “which button do I press?” pause on every visit that drains goodwill.\nRecommendation: 1A because DESIGN.md already names the exact tokens and it keeps one accent color on the page (Hierarchy as service; Use conventions).\nCompleteness: 1A=10/10, 1B=8/10, 1C=2/10\nNet: one filled button versus a second tier that DESIGN.md does not define, versus leaving the page with no primary action.": "1A) DESIGN.md primary (recommended)"
+      },
+      "unansweredQuestionIndices": [],
+      "answeredAt": "2026-09-21T18:09:29.711Z"
+    }
+  ]
+}