docs: D202 M-63/M-65/M-66/M-66b + D204 writeups; audio status in README/AGENTS

- findings.md: D202 full session history (M-63 live repro, M-65 voice-leak
  guard + sndCreatePostEvent unstub, M-66 root cause — sound 203 infinite
  ADPCM loop played fire-and-forget = faithful N64 quirk; M-66b option-C
  expiration implemented + measured) and D204 (AI-queue starvation, fixed
  + measured). Index row updated; findings-index.csv regenerated.
- GE-ENV-PROBES.md: D202 probe set documented ([EXPIRE] marked permanent).
- GRAPHICS-BACKLOG.md: D172 / D176 / 1P-viewmodel rows point at the new
  research docs.
- README + AGENTS.md: audio status corrected (mixer landed post-v0.1.0;
  Phase 3 in progress, not 'no audio').

AGENTS.md

@@ -56,6 +56,8 @@ Run `/linkcheck` for this sweep. Record new findings in `docs/dev/findings.md`
   parked in `docs/dev/GRAPHICS-BACKLOG.md`.
 - **Phase 3 (audio + input):** input layer done (`port/src/input.c`); polish
   bugs open (D118* mouse-look residuals; interactive feel-checks owed). Audio
-  not started — libaudio → SDL, copy-and-adapt from the PD port's `mixer.c`.
+  mixer done (libaudio → SDL software mixer, D198–D201); D204 tempo drift
+  fixed + measured; D202 stuck door loop root-caused with a port-side
+  expiration (M-66b) awaiting by-ear verification.
 - **Phase 4 (saves + polish):** file-backed EEPROM done; widescreen, config,
   rebinding UI outstanding.