Add bounded CPU VAD diagnostic with real model contract tests

docs/ROADMAP.md

@@ -79,6 +79,9 @@ release. Keep their evidence and unchecked production gates intact.
   calibrated handling of queries without a relevant match.
   - [x] Prepare bounded CPU speech diagnostics and test process failures,
     timestamp units and frozen-label scoring; see [speech proposal](SPEECH-ALIGNMENT.md).
+  - [x] Build a pinned CPU VAD diagnostic with explicit one-thread inference,
+    native error detection and full probabilities; actual Silero silence/input
+    contracts passed. Media quality comparison remains open (see SPEECH-VAD.md).
   - [ ] Run genuine local speech candidates, Google transcript retrieval and
     held-out English/German/music/no-speech checks before selecting a default.
 - [ ] Package and verify decoders on every release OS; test production acceleration.

docs/SPEECH-VAD.md

@@ -1,8 +1,9 @@
 # Speech gating and timestamp research
 
 Research recorded 2026-10-13 against whisper.cpp **v1.9.5**, exact commit
-`d1be6fde11ac6e0407606b4e42fe72d34add8037`. No VAD inference, alignment
-experiment, product integration or default change has been performed. The
+`d1be6fde11ac6e0407606b4e42fe72d34add8037`. A real-model CPU silence contract
+check has now passed; no media comparison, alignment experiment, product
+integration or default change has been performed. The
 corrected ungated ASR observations remain in [speech alignment](SPEECH-ALIGNMENT.md).
 The instrumental piano's large CPU cost and unrelated text motivate testing;
 they do not establish that any particular VAD model fixes the problem.
@@ -171,6 +172,17 @@ Source review identified further requirements before accepting measurements:
   mode and one thread, check segment allocation, retain probabilities and
   separate model loading, detection and segmentation timings.
 
+The separately tracked [diagnostic wrapper](../tools/speech-vad/README.md) also
+built successfully. Its SHA-256 is
+`095a05c85c6f8499608ba8caa46b7dc3bbc0258a8585eeecec74ccee4b3e19d4`.
+Three native contract tests passed in 1.451 seconds using the verified v6.2.0
+model: 16,001 silent samples yielded all 32 expected probabilities and no speech
+intervals; empty/misaligned/nonfinite/out-of-range PCM and unknown arguments
+failed without JSON. The wrapper fixes CPU mode, one inference thread and the
+parameter table, catches native error-level logs, and checks probability frame
+coverage. This is a smoke test, not a speech-quality result or OS thread-count
+measurement. An independent complete media comparison remains necessary.
+
 Open gates: actual one-thread audit, detection/recognition
 comparison, ordinary and DTW timestamp regression, independent held-out quality,
 background starvation/cancellation/crash recovery, packaging/license review,