feat(backend): scan-product accuracy 66.2% -> 79.7% + frozen validation benchmark

Accuracy work on the 79-image product-scan validation set (user goal: 90%):
- classify_ocr_server.py: 0/90/180/270-degree expiry-date search (stops at
  first hit, 0-degree fallback); classification decoupled onto the upright
  image (rotated frames regressed DINOv2 -6pts until this); cross-line date
  stitching; tiled full-res OCR pass (defeats the 4000px downscale that
  killed small inkjet dates); VL-pipeline expiry fallback with
  keyword-anchored anti-hallucination guard; VL text lines merged into
  text_lines + VL SKU retry. Visualization endpoints removed entirely
  (Visual/Spotting grids - unused by frontend, 3x per-scan GPU cost).
- product-scan.ts: coverage-normalized OCR-evidence re-ranking of DINOv2
  top-K (tuned offline: +8/-0 on top-1 misses), re-ranked class mapped to
  sku_master by SKU prefix; classifier timeout 90s->240s for fallback paths.
- Frozen benchmark: product-test-images-fixed/ (79 renamed images) +
  freeze/seed/build-undetected/capture/experiment scripts; labels trimmed to
  the 79 validation entries (training rows kept in .bak-with-training);
  5 TRAINED-ON SKUs replaced with fresh held-out photos.
- manual-label-scan page: shows last batch-test AI prediction under every
  field by default (new /api/product-scan-results); serves the fixed folder;
  fixed total hydration failure via allowedDevOrigins 127.0.0.1.
- Measured (all-79, zero failures): sku/name 87.3%, expiry 64.6%, overall
  79.7%. Tiles/VL-evidence/VL-SKU deployed but not yet batch-measured.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Gr6HH7JrdsXX8AARejQboM

backend/sources/product-test-images/README.md

@@ -1,18 +1,22 @@
-# Product-scan validation images
+# Product-scan validation images — live intake (staging)
 
-This folder is the **validation/test set** for the product-scan accuracy
-harness (`backend/scripts/accuracy-check-scan.mts`) — real-world photos that
-are *not* part of the classifier's reference dataset, so scoring against them
-measures actual accuracy instead of memorization.
+This folder is the **live-intake / staging area** for the product-scan
+validation set. It's still where the `/manual-label-scan` page saves new
+photo drops, and still what real-world photos get dropped into by hand — but
+it is **no longer what the accuracy harness scores**. That's
+`../product-test-images-fixed/` (a frozen, sequentially-renamed snapshot) —
+see that folder's README for why the split exists.
 
-**Workflow:**
+**Workflow (adding a new SKU or photo):**
 1. Drop a photo here directly (flat, no subfolders — a filename with no `/`
    is what marks an image as "validation" instead of "training").
 2. Label it via the `/manual-label-scan` page (correct `no_sku`, `nama_item`,
    `expiry_date` by hand — don't just accept the AI-scan prefill, that would
    make the ground truth equal to the model's own prediction).
-3. Run `node scripts/accuracy-check-scan.mts` from `backend/` — the photo now
-   scores under "Validation Set", separate from "Training Set".
+3. Re-run `node scripts/freeze-validation-set.mjs` from `backend/` to promote
+   the new photo into `product-test-images-fixed/` (renamed to
+   `<index> <no_sku>.<ext>`) so it actually gets scored on the next
+   `accuracy-check-scan.mts` run.
 
 **This is not where new training photos go.** To improve the classifier
 itself (DINOv2 index / YOLO fine-tune), add photos to