docs: P3.5 mix/ducking review findings (calibration deferred to P3.6)

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Bg76sLc43Wc3j5ZcLkboYR

docs/cinematic-highlight-poc-plan.md

@@ -92,7 +92,14 @@ Ranked from the first render's evidence:
       (committed localpoc profile keeps the safe heuristic default; worker started manually).
       CAVEATS: yolov8n.pt is **AGPL-3.0** (production blocker on this license alone; see yolov8n.pt.license.txt);
       loopback HTTP is flagged non-compliant for a certified path; Haar face detector gives false positives.
-- [ ] P3.5 Music/SFX creative fit + ducking review; consider longer inference / better prompts.
+- [~] P3.5 Music/SFX + ducking review (measurement-based). FINDINGS: sidechain ducking IS implemented
+      (`[music_raw][voice]sidechaincompress` with the configured threshold/ratio/attack/release; FFmpeg
+      auto-splits `[voice]` so it both keys the duck and stays in the mix) and it fires (music drops during
+      voice). Source audio is near-silent (-54 dB RMS) -> negligible. Loudness/true-peak stay in spec.
+      OPEN: voice-vs-music balance cannot be validated or tuned by measurement alone — MusicGen produces
+      different audio each run, so cross-render A/B is confounded, and this needs LISTENING. A blind +3.5 dB
+      voice boost was tried and reverted (could not verify it helped; measurement suggested it did not).
+      Per the "do not tune audio blindly" discipline, mix calibration is deferred to P3.6 (human review).
 - [ ] P3.6 Freeze acceptance thresholds + blinded human creative review vs baseline before declaring success.
 
 ## Milestone log