Fix IPA correction for dictionary pages (WIP)

- Fix Step 5h: restrict slash-IPA conversion to English headword column
  only — prevents converting "der/die/das" to "der [dər]das" in German
  columns (confirmed working)
- Fix _text_has_garbled_ipa: detect embedded apostrophes in merged
  tokens like "Scotland'skotland" where OCR reads ˈ as '
- Fix _insert_missing_ipa: detect dictionary word prefix in merged
  trailing tokens like "fictionsalans'fIkfn" → extract "fiction" with IPA
- Move en_col_type to wider scope for Step 5h access

Note: Fixes 1+2 confirmed working in unit tests but not yet applying
in the full build-grid pipeline — needs further debugging.

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

klausur-service/backend/grid_editor_api.py

@@ -858,6 +858,7 @@ async def _build_grid_core(session_id: str, session: dict) -> dict:
     # Single/two-column layouts are continuous text, not vocab tables.
     all_cells = [cell for z in zones_data for cell in z.get("cells", [])]
     total_cols = sum(len(z.get("columns", [])) for z in zones_data)
+    en_col_type = None
     if total_cols >= 3:
         # Find the column that contains IPA brackets → English headwords.
         # Count cells with bracket patterns per col_type.  The column with
@@ -872,7 +873,6 @@ async def _build_grid_core(session_id: str, session: dict) -> dict:
             if ct.startswith("column_") and '[' in txt:
                 col_bracket_count[ct] = col_bracket_count.get(ct, 0) + 1
         # Pick column with most bracket IPA patterns
-        en_col_type = None
         if col_bracket_count:
             en_col_type = max(col_bracket_count, key=col_bracket_count.get)
         else:
@@ -1105,6 +1105,10 @@ async def _build_grid_core(session_id: str, session: dict) -> dict:
     slash_ipa_fixed = 0
     for z in zones_data:
         for cell in z.get("cells", []):
+            # Only process English headword column — avoid converting
+            # German text like "der/die/das" to IPA.
+            if en_col_type and cell.get("col_type") != en_col_type:
+                continue
             text = cell.get("text", "")
             if "/" not in text:
                 continue