fix: _split_broad_columns nur bei maximal 1 breiter Spalte ausfuehren

Wenn bereits 2+ breite Content-Spalten existieren, ist das Layout
wahrscheinlich korrekt in EN/DE getrennt. Split wird nur ausgefuehrt
wenn eine einzelne breite Spalte EN+DE kombiniert enthaelt.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

klausur-service/backend/cv_vocab_pipeline.py

@@ -2097,6 +2097,14 @@ def _split_broad_columns(
     logger.info(f"SplitBroadCols: input {len(geometries)} cols: "
                 f"{[(g.index, g.x, g.width, g.word_count, round(g.width_ratio, 3)) for g in geometries]}")
 
+    # Count how many broad content columns exist.  If there are already 2+,
+    # the layout is likely already correctly split into EN / DE — skip.
+    broad_count = sum(1 for g in geometries
+                      if g.width_ratio > _broad_threshold and len(g.words) >= 10)
+    if broad_count >= 2:
+        logger.info(f"SplitBroadCols: {broad_count} broad cols already → skip")
+        return geometries
+
     for geo in geometries:
         if geo.width_ratio <= _broad_threshold or len(geo.words) < 10:
             result.append(geo)