lipsync: keep the managed fallback out of the packaged install's directory

Correction to the previous commit. The OCI image already ships a complete
lip-sync integration — packaging/linux/launcher/{wav2lip,sadtalker} shims on
PATH, a dedicated lipsync_venv, baked Wav2Lip AND SadTalker source, first-use
checkpoint download and a smoke test. The claim that the binary "essentially
never exists" was wrong for the packaged deployment; _apply_lipsync() uses those
shims and never reaches the managed path there.

The managed install is therefore only for from-source installs (build.sh +
venv_all), where nothing is on PATH and lip sync did silently degrade.

More importantly it had claimed CODERAI_WAV2LIP_DIR — the same variable the shim
uses for its writable working copy, seeded from the baked read-only source and
run against the lipsync venv's pinned deps. Applying our numpy/librosa/torch
rewrites to that tree would have corrupted a working packaged install. The
managed copy now lives under CODERAI_WAV2LIP_MANAGED_DIR
(~/.coderai/wav2lip-managed), whose basename also cannot collide with the shim's
~/.coderai/Wav2Lip on a case-insensitive filesystem.

Re-verified after the move: fetch/patch/run still produces a valid synced mp4.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01R6iugoNeEgt9PkajKrFyq4

README.md

@@ -114,7 +114,7 @@ Full guide: [`docs/runpod.md`](docs/runpod.md).
 - **Character Profiles**: Apply up to 6 saved character profiles for visual consistency. Identity is conditioned via IP-Adapter on pipelines that support it (AnimateDiff and other SD-based video models); pipelines without IP-Adapter (Wan, CogVideoX, LTX, SVD) fall back to the prompt hint plus any per-character LoRA, and say so in the log
 - **Environment Profiles**: Apply up to 6 saved environment profiles to condition the scene
 - **Multi-Character Dialog**: Assign spoken lines to individual characters — each line picks a character profile, voice profile or TTS voice ID, and text; lines are **voice-cloned with F5-TTS** from the named profile, mixed with correct timing (sequential or manual), and lip-synced via Wav2Lip
-- **Lip Sync**: Wav2Lip is a **managed install** — the code and weights are fetched and patched on first use, so no manual setup. If it can't run, the response carries a `warnings` entry saying the audio was muxed without mouth sync rather than passing off an unsynced video as a success
+- **Lip Sync**: Wav2Lip and SadTalker. The Docker/OCI image ships both (launcher shims plus a dedicated lip-sync venv; checkpoints download on first use). A from-source install has no shims, so CoderAI falls back to a self-managed Wav2Lip that fetches and patches itself on first use. Either way, if lip sync can't run the response carries a `warnings` entry saying the audio was muxed without mouth sync — rather than passing off an unsynced video as a success
 
 ### Audio
 - **Text-to-Speech**: Kokoro TTS with voice selection and speed control

codai/api/lipsync.py

@@ -6,18 +6,24 @@
 # the Free Software Foundation, either version 3 of the License, or
 # (at your option) any later version.
 
-"""Managed Wav2Lip lip-sync.
-
-Wav2Lip has no pip package and no installable binary; upstream expects you to clone
-the repo and hunt down checkpoints whose original links have long since rotted. So
-this module owns the whole lifecycle the way :mod:`codai.api.ds4_worker` owns ds4:
-it fetches the code *and* the weights from a HuggingFace mirror on first use, patches
-the parts that no longer run on modern numpy/librosa/torch, and drives ``inference.py``
-as a subprocess.
-
-That replaces the previous behaviour, which looked for a ``wav2lip`` executable on
-PATH — a file that essentially never exists — and silently returned an unsynchronised
-video when it was missing.
+"""Self-contained Wav2Lip lip-sync for installs that don't ship the packaged shims.
+
+**The OCI image already handles lip sync properly** — ``packaging/linux/launcher/``
+installs ``wav2lip`` and ``sadtalker`` shims on PATH which run the baked repo code in a
+dedicated ``lipsync_venv`` and download the checkpoints on first use. When those shims
+are present :func:`codai.api.video._apply_lipsync` uses them and never reaches this
+module.
+
+This module covers the *other* case: a from-source install (``build.sh`` + ``venv_all``)
+where nothing is on PATH, which is where lip sync silently degraded to a plain audio mux.
+It owns the lifecycle the way :mod:`codai.api.ds4_worker` owns ds4 — fetching code *and*
+weights from a HuggingFace mirror on first use, patching what no longer runs on modern
+numpy/librosa/torch, and driving ``inference.py`` as a subprocess.
+
+It deliberately uses its **own** directory (``CODERAI_WAV2LIP_MANAGED_DIR``) rather than
+the packaged ``CODERAI_WAV2LIP_DIR``: that one is the shim's writable working copy, seeded
+from the baked read-only source and run against the lipsync venv's pinned dependencies.
+Patching it from here would corrupt a working packaged install.
 """
 
 import os
@@ -43,8 +49,11 @@ _ready: Optional[Tuple[Path, Path]] = None
 
 
 def default_install_dir() -> Path:
-    return Path(os.environ.get("CODERAI_WAV2LIP_DIR")
-                or os.path.expanduser("~/.coderai/wav2lip"))
+    # NOT CODERAI_WAV2LIP_DIR — that belongs to the packaged launcher shim. The
+    # distinct basename also avoids colliding with the shim's ~/.coderai/Wav2Lip on a
+    # case-insensitive filesystem.
+    return Path(os.environ.get("CODERAI_WAV2LIP_MANAGED_DIR")
+                or os.path.expanduser("~/.coderai/wav2lip-managed"))
 
 
 # --------------------------------------------------------------------------- #
@@ -131,7 +140,7 @@ def ensure_wav2lip(install_dir: Optional[Path] = None) -> Tuple[Path, Path]:
             raise RuntimeError(
                 f"could not download Wav2Lip assets from {WAV2LIP_REPO}: {exc}. "
                 f"Place the repo and checkpoints/wav2lip_gan.pth in {root} manually, "
-                f"or set CODERAI_WAV2LIP_DIR.") from exc
+                f"or set CODERAI_WAV2LIP_MANAGED_DIR.") from exc
 
     if not (root / "inference.py").exists():
         raise RuntimeError(f"Wav2Lip inference.py missing under {root}")