fix(rectification): narrate exhausted-probe adopt instead of keep-collecting

When leftover probes cannot split adjacent minutes, skip frameless follow-ups and let a no-tool agent explain the stop. Distinguish-card "no" no longer closes a whole evidence domain.

Co-authored-by: Cursor <cursoragent@cursor.com>

frontend/tests/rectification-answer-choice.test.ts

@@ -1042,8 +1042,13 @@ test("failed card persist still leaves a spoken collect next step", () => {
   assert.doesNotMatch(source, /if \(fallback\.status === "created"/);
   const applied = source.slice(source.indexOf("async function persistApplied"));
   const loadAt = applied.indexOf("loadV9CaseCompute");
-  const decideAt = applied.indexOf("publicNextAction(decideAfterInferenceChange");
+  // 原值: publicNextAction(decideAfterInferenceChange({...})) 内联一次。
+  // 新值: 先 decideAfterInferenceChange 得到 nextDecision，再 publicNextAction(nextDecision)。
+  // 原因: 采用旁白需要完整 decision（precisionStage / droppedProbes），不能只拿公开 nextAction。
+  const decideAt = applied.indexOf("decideAfterInferenceChange({");
+  const publicAt = applied.indexOf("publicNextAction(nextDecision)");
   assert.ok(loadAt >= 0 && decideAt >= 0 && loadAt < decideAt);
+  assert.ok(publicAt > decideAt);
   assert.match(applied.slice(decideAt, decideAt + 280), /birthDate/);
   assert.match(
     applied.slice(applied.indexOf("persistNextInterviewAfterChoice"), applied.indexOf("persistNextInterviewAfterChoice") + 360),