fix(ocr): reduce memory usage for 16GB dev machines

- Surya models lazy-load on first OCR request instead of at startup
  (saves ~3-4GB idle RAM — Kraken stays eager at ~16MB)
- Process one page at a time in Surya engine (limits peak memory)
- RECOGNITION_BATCH_SIZE=1, DETECTOR_BATCH_SIZE=1 (slower but fits in RAM)
- Revert mem_limit back to 6GB (sufficient with these optimizations)
- Render DPI stays at 200

Idle memory: ~2GB (Kraken only). Peak during OCR: ~5-6GB (Surya loaded).

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

docker-compose.yml

@@ -78,14 +78,16 @@ services:
       dockerfile: Dockerfile
     container_name: archive-ocr
     restart: unless-stopped
-    mem_limit: 10g
-    memswap_limit: 10g
+    mem_limit: 6g
+    memswap_limit: 6g
     volumes:
       - ocr_models:/app/models
     environment:
       KRAKEN_MODEL_PATH: /app/models/german_kurrent.mlmodel
       OCR_CONFIDENCE_THRESHOLD: "0.3"
       OCR_CONFIDENCE_THRESHOLD_KURRENT: "0.5"
+      RECOGNITION_BATCH_SIZE: "1"
+      DETECTOR_BATCH_SIZE: "1"
     networks:
       - archive-net
     healthcheck: