feat: add Van's real baseline lane and stop size tokens naming products

Two review findings, both confirmed by measurement.

1. The evaluator had no always-argmax lane, so every "vs Van" claim compared
   against the *gated* matcher, which is not Van's contract - the original demo
   answered every crop with its best-scoring SKU and had no accept line. The new
   'van' lane is that contract: same scores, no gate. Against it the local lane
   wins both axes at both resolutions (full 61%/77% vs 52%/29%; live 59%/95% vs
   57%/32%), but the live coverage margin is one facing, not seven, and the 95%
   figure is live-resolution only - at full resolution the same lane reads 77%
   with 11 false accepts and a hub alarm.

2. A crop reading only "400g" scored 0.667 on three SKUs, above TEXT_ACCEPT, so
   the text lane named a product from a size line. The fill-in job now requires
   two non-digit bigrams; the veto is unchanged, so "65%" still contradicts a
   local answer. The raw OCR read travels with the ranking through
   score_crop_lanes for that check.

Tests: 3 for the size gate (policy + live) and the van lane's own row. Removing
the gate fails all three. Suite 276 -> 280.

recognization/code/identity_lanes.py

@@ -263,6 +263,31 @@ def bigrams(text: str) -> set:
     return {text[i:i + 2] for i in range(len(text) - 1)}
 
 
+def identity_bigrams(text: str) -> set:
+    """Bigrams carrying a non-digit character.
+
+    A crop reading only "400g" matches every 400g product's size line, so size
+    and number tokens must not name a product on their own. They still count
+    for the veto, where the local lane has already supplied the evidence and
+    the text only has to contradict it.
+    """
+    return {gram for gram in bigrams(text) if not gram.isdigit()}
+
+
+
+def carries_identity(read: str) -> bool:
+    """Whether a read holds enough non-numeric text to name a product.
+
+    A crop reading only "400g" matches every 400g product's size line, so size
+    and number tokens must not answer on their own. Two non-digit bigrams is
+    the floor: "400g" has one ("0g") and is refused, while "コクうま 65%" has
+    five and still names l09. The veto is unaffected - there the local lane has
+    already supplied the evidence and the text only has to contradict it, so
+    "65%" stays usable.
+    """
+    return len(identity_bigrams(normalize_text(read))) >= 2
+
+
 def text_similarity(read: str, signature: str) -> float:
     """Share of the read text's bigrams that the signature explains.
 
@@ -352,6 +377,18 @@ class TextLane:
         """Bigram similarity per catalog row. Empty when nothing was read."""
         return self.scores_from_text(self.read(crop))
 
+    def carries_identity(self, read: str) -> bool:
+        """Whether a read holds enough non-numeric text to name a product.
+
+        A crop reading only "400g" matches every 400g product's size line, so
+        size and number tokens must not answer on their own. Two non-digit
+        bigrams is the floor: "400g" has one ("0g") and is refused, while
+        "コクうま 65%" has five and still names l09. The veto is unaffected -
+        there the local lane has already supplied the evidence and the text
+        only has to contradict it, so "65%" stays usable.
+        """
+        return len(identity_bigrams(normalize_text(read))) >= 2
+
     def scores_from_text(self, read: str) -> Dict[str, float]:
         if len(normalize_text(read)) < 4:
             return {}
@@ -412,8 +449,14 @@ class DecisionPolicy:
         text_rank: Any = None,
         *,
         details: Optional[Dict[str, Any]] = None,
+        text_read: Optional[str] = None,
     ) -> Dict[str, Any]:
-        """One decision, in the shape `decide_identity` already returns."""
+        """One decision, in the shape `decide_identity` already returns.
+
+        `text_read` is the raw OCR text behind `text_rank`. It gates the
+        fill-in job only: a read that is nothing but a size or a number cannot
+        name a product, but it can still veto one.
+        """
         from identity import (
             IDENTITY_ACCEPTED,
             IDENTITY_AMBIGUOUS,
@@ -430,6 +473,7 @@ class DecisionPolicy:
             if text_top
             and text_top[1] >= self.text_accept
             and (text_top[1] - text_runner) >= self.text_margin
+            and (text_read is None or carries_identity(text_read))
             else None
         )