feat: Words-First Grid Builder (bottom-up alternative zu cell_grid_v2)

Neuer Algorithmus in cv_words_first.py: Clustert Tesseract word_boxes
direkt zu Spalten (X-Gap) und Zeilen (Y-Proximity), baut Zellen an
Schnittpunkten. Kein Spalten-/Zeilenerkennung noetig.

- cv_words_first.py: _cluster_columns, _cluster_rows, _build_cells, build_grid_from_words
- ocr_pipeline_api.py: grid_method Parameter (v2|words_first) im /words Endpoint
- StepWordRecognition.tsx: Dropdown Toggle fuer Grid-Methode
- OCR-Pipeline.md: Doku v4.3.0 mit Words-First Algorithmus
- 15 Unit-Tests fuer cv_words_first

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

admin-lehrer/components/ocr-pipeline/StepWordRecognition.tsx

@@ -63,6 +63,7 @@ export function StepWordRecognition({ sessionId, onNext, goToStep, skipHealGaps
   const [ocrEngine, setOcrEngine] = useState<'auto' | 'tesseract' | 'rapid'>('auto')
   const [usedEngine, setUsedEngine] = useState<string>('')
   const [pronunciation, setPronunciation] = useState<'british' | 'american'>('british')
+  const [gridMethod, setGridMethod] = useState<'v2' | 'words_first'>('v2')
 
   // Streaming progress state
   const [streamProgress, setStreamProgress] = useState<{ current: number; total: number } | null>(null)
@@ -112,7 +113,7 @@ export function StepWordRecognition({ sessionId, onNext, goToStep, skipHealGaps
       let res: Response | null = null
       for (let attempt = 0; attempt < 2; attempt++) {
         res = await fetch(
-          `${KLAUSUR_API}/api/v1/ocr-pipeline/sessions/${sessionId}/words?stream=true&engine=${eng}&pronunciation=${pronunciation}${skipHealGaps ? '&skip_heal_gaps=true' : ''}`,
+          `${KLAUSUR_API}/api/v1/ocr-pipeline/sessions/${sessionId}/words?stream=${gridMethod === 'v2' ? 'true' : 'false'}&engine=${eng}&pronunciation=${pronunciation}${skipHealGaps ? '&skip_heal_gaps=true' : ''}&grid_method=${gridMethod}`,
           { method: 'POST' },
         )
         if (res.ok) break
@@ -128,6 +129,13 @@ export function StepWordRecognition({ sessionId, onNext, goToStep, skipHealGaps
         throw new Error(err.detail || 'Worterkennung fehlgeschlagen')
       }
 
+      // words_first returns plain JSON (no streaming)
+      if (gridMethod === 'words_first') {
+        const data = await res.json() as GridResult
+        applyGridResult(data)
+        return
+      }
+
       const reader = res.body!.getReader()
       const decoder = new TextDecoder()
       let buffer = ''
@@ -220,7 +228,7 @@ export function StepWordRecognition({ sessionId, onNext, goToStep, skipHealGaps
       setDetecting(false)
     }
   // eslint-disable-next-line react-hooks/exhaustive-deps
-  }, [sessionId, ocrEngine, pronunciation])
+  }, [sessionId, ocrEngine, pronunciation, gridMethod])
 
   const handleGroundTruth = useCallback(async (isCorrect: boolean) => {
     if (!sessionId) return
@@ -789,6 +797,16 @@ export function StepWordRecognition({ sessionId, onNext, goToStep, skipHealGaps
       {gridResult && (
         <div className="bg-white dark:bg-gray-800 rounded-xl border border-gray-200 dark:border-gray-700 p-4 space-y-3">
           <div className="flex items-center gap-3 flex-wrap">
+            {/* Grid method selector */}
+            <select
+              value={gridMethod}
+              onChange={(e) => setGridMethod(e.target.value as 'v2' | 'words_first')}
+              className="px-2 py-1.5 text-xs border rounded-lg dark:bg-gray-700 dark:border-gray-600"
+            >
+              <option value="v2">Standard (v2)</option>
+              <option value="words_first">Words-First</option>
+            </select>
+
             {/* OCR Engine selector */}
             <select
               value={ocrEngine}