fix: model weights setup — export scripts, ORT compat, bootstrap

- Add export_models.py for OpenCLIP ViT-B/32 and YOLOv8n ONNX export
- Fix ArgMax(13) ORT ARM64 incompatibility by passing eot_indices as a
  separate ONNX input (computed outside the graph in embed.py)
- Use legacy TorchScript exporter (dynamo=False) for IR version 9 compat
- Upgrade onnxruntime to 1.18.1
- Rewrite bootstrap_models.py with clear separation of auto-downloadable
  models (YuNet, SFace) vs manually-exported ones (OpenCLIP, YOLOv8n)
- Wire bootstrap into worker CMD (runs before Celery)

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

backend/app/services/vision/bootstrap_models.py

@@ -4,7 +4,11 @@ Download vision model weights on first worker boot.
 Run as:  python -m app.services.vision.bootstrap_models
 
 Or called from the vision worker entrypoint before Celery starts.
-Downloads are idempotent — existing files with matching sizes are skipped.
+Downloads are idempotent — existing files are skipped.
+
+For models that require export (OpenCLIP, YOLOv8n), see export_models.py.
+Those must be exported once on any machine with pip, then placed in
+the models volume before the worker starts.
 """
 import logging
 import os
@@ -15,35 +19,30 @@ from app.config import settings
 
 logger = logging.getLogger(__name__)
 
-# (relative_path, url, expected_size_bytes_approx)
-# Sizes are approximate — used only for skip-if-exists checks, not integrity.
-MODELS = [
-    # OpenCLIP ViT-B/32 — visual and textual encoders (ONNX)
-    # These must be exported manually via export_openclip.py (see below).
-    # Placeholder entries — bootstrap will warn if missing.
-    ("embed/visual.onnx", None, None),
-    ("embed/textual.onnx", None, None),
-
-    # YOLOv8n — object detection
-    # Export: `yolo export model=yolov8n.pt format=onnx imgsz=640`
-    # Placeholder — must be exported from ultralytics offline.
-    ("detect/yolov8n.onnx", None, None),
-
-    # YuNet — face detection (Apache 2.0, opencv_zoo)
+# (relative_path, url, description)
+# Models with url=None must be pre-exported via export_models.py.
+DOWNLOADS = [
+    # YuNet — face detection (Apache 2.0, opencv_zoo, ~233 KB)
     (
         "face/yunet.onnx",
         "https://github.com/opencv/opencv_zoo/raw/main/models/face_detection_yunet/face_detection_yunet_2023mar.onnx",
-        233_000,
+        "YuNet face detector",
     ),
-
-    # SFace — face recognition (Apache 2.0, opencv_zoo)
+    # SFace — face recognition (Apache 2.0, opencv_zoo, ~37 MB)
     (
         "face/sface.onnx",
         "https://github.com/opencv/opencv_zoo/raw/main/models/face_recognition_sface/face_recognition_sface_2021dec.onnx",
-        37_000_000,
+        "SFace face recognizer",
     ),
 ]
 
+# Models that need manual export via export_models.py
+EXPORTS = [
+    ("embed/visual.onnx", "OpenCLIP ViT-B/32 visual encoder"),
+    ("embed/textual.onnx", "OpenCLIP ViT-B/32 textual encoder"),
+    ("detect/yolov8n.onnx", "YOLOv8n object detector"),
+]
+
 
 def bootstrap(models_dir: str | None = None):
     """Ensure all model files are present. Download what we can, warn about
@@ -51,32 +50,46 @@ def bootstrap(models_dir: str | None = None):
     base = Path(models_dir or settings.vision.models_dir)
     base.mkdir(parents=True, exist_ok=True)
 
-    for rel_path, url, expected_size in MODELS:
+    # Download auto-downloadable models
+    for rel_path, url, desc in DOWNLOADS:
         dest = base / rel_path
         dest.parent.mkdir(parents=True, exist_ok=True)
 
         if dest.exists():
-            logger.debug("Model already exists: %s", dest)
+            logger.debug("Already exists: %s (%s)", dest, desc)
             continue
 
-        if url is None:
-            logger.warning(
-                "Model file %s not found and has no auto-download URL. "
-                "See bootstrap_models.py for export instructions.",
-                dest,
-            )
-            continue
-
-        logger.info("Downloading %s → %s", url, dest)
+        logger.info("Downloading %s → %s", desc, dest)
         try:
             urlretrieve(url, str(dest))
-            actual = dest.stat().st_size
-            logger.info("Downloaded %s (%d bytes)", rel_path, actual)
+            size_kb = dest.stat().st_size / 1024
+            logger.info("Downloaded %s (%.0f KB)", desc, size_kb)
         except Exception as e:
-            logger.error("Failed to download %s: %s", rel_path, e)
+            logger.error("Failed to download %s: %s", desc, e)
             if dest.exists():
                 dest.unlink()
 
+    # Check for manually-exported models
+    missing = []
+    for rel_path, desc in EXPORTS:
+        dest = base / rel_path
+        if not dest.exists():
+            missing.append((rel_path, desc))
+
+    if missing:
+        logger.warning(
+            "Missing %d model(s) that require manual export via export_models.py:",
+            len(missing),
+        )
+        for rel_path, desc in missing:
+            logger.warning("  %s — %s", base / rel_path, desc)
+        logger.warning(
+            "Run: python -m app.services.vision.export_models --models-dir %s",
+            base,
+        )
+    else:
+        logger.info("All model files present in %s", base)
+
 
 if __name__ == "__main__":
     logging.basicConfig(level=logging.INFO)