Fix false edge-whitening on die-cut/mixed-material cards (e.g. SPx clear acetate window)

The pooled baseline that scores edge whitening was global across all four
edges, so a card that's normal printed border on three sides and clear
acetate on the fourth had the three normal edges anchor a baseline that made
the clear side read as extreme whitening -- correctly detecting a real pixel
difference, just the wrong one. Now excludes an edge from both the pool and
its own scoring when its median brightness/saturation reads as a
fundamentally different material, and surfaces why in the prompt instead of
silently dropping it.

cardimage.py

@@ -415,6 +415,13 @@ def edge_wear_profile(image_bytes):
     and which are immune to the thing that kept defeating it: telling a
     genuine pale band apart from the cut line and the border's own
     anti-aliasing.
+
+    An individual edge can come back None with an entry in edge_notes rather
+    than a score, when that edge's own material reads as fundamentally
+    different from the rest of the card's border (a die-cut clear window, a
+    foil accent strip on one side only) — see the outlier detection below for
+    why scoring it against the other edges' baseline would be actively wrong,
+    not just imprecise.
     """
     if Image is None:
         return None
@@ -450,15 +457,62 @@ def edge_wear_profile(image_bytes):
         if all(v is None for v in collected.values()):
             return None
 
-        # Baseline from ALL four edges pooled, not each edge against itself.
+        # Before pooling, catch an edge whose FINISH differs from the rest of
-        # Whitening only ever raises luma and lowers saturation, so unworn
+        # the card outright — a die-cut window of clear acetate, a foil
-        # border sits at the low end of one and the high end of the other,
+        # accent strip on one side only — rather than one that's simply worn.
-        # and quartiles across the whole card find it. Scoring an edge
+        # This is different from the whole-card foil/silver check below: that
+        # one catches a card that's uniformly pale everywhere, but a die-cut
+        # insert is normal printed border on three sides and something else
+        # entirely on the fourth, so the whole-card check never trips — the
+        # three normal edges keep the pooled baseline looking sane, which is
+        # exactly what then makes the fourth edge look catastrophically
+        # whitened. Wear doesn't produce this: even a badly frayed edge is
+        # still mostly the same border material with patches of paper
+        # showing through, so its median luma/saturation barely moves. A
+        # genuinely different material moves the median far more than
+        # ordinary wear or lighting ever does.
+        edge_medians = {}
+        for name, cols in collected.items():
+            if not cols:
+                continue
+            ls = sorted(c[0] for c in cols)
+            ss = sorted(c[1] for c in cols)
+            edge_medians[name] = (ls[len(ls) // 2], ss[len(ss) // 2])
+
+        outliers = {}
+        for name, (l_med, s_med) in edge_medians.items():
+            others = [v for n, v in edge_medians.items() if n != name]
+            if len(others) < 2:
+                continue
+            other_l = sorted(v[0] for v in others)[len(others) // 2]
+            other_s = sorted(v[1] for v in others)[len(others) // 2]
+            if (l_med - other_l) >= 45 and (other_s - s_med) >= 35:
+                outliers[name] = (
+                    "this edge's finish reads as a different material from "
+                    "the card's other edges (much brighter and less "
+                    "saturated) — likely a clear acetate window, a die-cut "
+                    "insert, or a foil accent on this side only, not "
+                    "whitening. A paper-showing-through measurement doesn't "
+                    "apply to a material that was never opaque to begin with."
+                )
+
+        # Baseline from the non-outlier edges pooled, not each edge against
+        # itself. Whitening only ever raises luma and lowers saturation, so
+        # unworn border sits at the low end of one and the high end of the
+        # other, and quartiles across the pool find it. Scoring an edge
         # against only its own length silently fails on the case that
         # matters most — an edge worn evenly end to end, where the baseline
         # becomes the wear and the damage cancels itself out. Pooling means
-        # three clean edges anchor the fourth.
+        # the clean edges anchor the rest — an outlier edge left in this pool
+        # would drag the baseline toward itself and make the genuinely normal
+        # edges misread in turn, so it's excluded here.
+        pooled = [c for name, cols in collected.items()
+                  if cols and name not in outliers for c in cols]
+        if len(pooled) < 40:
+            # Nothing survived exclusion (or everything was already sparse) —
+            # fall back to the full pool rather than giving up outright.
             pooled = [c for cols in collected.values() if cols for c in cols]
+            outliers = {}
         if len(pooled) < 40:
             return None
         lumas = sorted(c[0] for c in pooled)
@@ -485,12 +539,16 @@ def edge_wear_profile(image_bytes):
                       "— typical of a refractor or prismatic finish — for a "
                       "whitening measurement to mean anything")
 
-        edges = {name: (_score_edge(cols, base_l, base_s) if cols else None)
+        edges = {
-                 for name, cols in collected.items()}
+            name: (None if name in outliers
+                   else (_score_edge(cols, base_l, base_s) if cols else None))
+            for name, cols in collected.items()
+        }
         return {
             "edges": edges,
             "reliable": reason is None,
             "reason": reason,
+            "edge_notes": outliers,
             "border_luma": round(base_l, 1),
             "border_saturation": round(base_s, 1),
         }

vision.py

@@ -676,11 +676,29 @@ def grade_card(images, api_key=None, model=None, effort=None, zoom_details=True)
                 measured.get("reason", "of its finish")))
     elif measured:
         rows = []
+        note_rows = []
+        edge_notes = measured.get("edge_notes") or {}
         for side in ("top", "right", "bottom", "left"):
             data = (measured.get("edges") or {}).get(side)
             if data:
                 rows.append("  {} edge: {:.1f}% of its length".format(
                     side, data["percent"]))
+            elif side in edge_notes:
+                note_rows.append("  {} edge: NOT MEASURED — {}".format(
+                    side, edge_notes[side]))
+        if note_rows:
+            prompt_parts.append(
+                "\n".join(note_rows) + "\n"
+                "For any edge noted above as not measured due to a different "
+                "material, judge it from its strip image alone, on its own "
+                "terms as whatever that material is — do not read its "
+                "brightness or lack of saturation as whitening just because "
+                "it looks nothing like the card's other edges. Look instead "
+                "for actual physical damage on that material: a crack or "
+                "chip in a clear acetate window, foil peeling or flaking, a "
+                "scuff that catches light differently than the surrounding "
+                "area — the same kind of judgement as any other edge, just "
+                "without a number to anchor it.")
         if rows:
             prompt_parts.append(
                 "MEASURED EDGE WHITENING — computed directly from the pixels, "