feat: RapidOCR default OCR engine, office-only ocr_engine, drop operator toggle

docs/architecture.md

@@ -110,7 +110,7 @@ stats panel (Loading / Unloading / Net / last-3-batch history), and a bottom sta
   (`models.modes`); missing file falls back to `.env` + legacy defaults.
 - `src/do_batch.py` — pure helpers for DO-gated batches (mode validation, plate
   grouping, start gates, net/discard both-nets rule, unit classify, retention).
-- `src/do_ocr.py` — `extract_do_fields(image, engine)` for `tesseract`|`paddle`|`none`
+- `src/do_ocr.py` — `extract_do_fields(image, engine)` for `rapid`|`tesseract`|`paddle`|`none` (default `rapid`)
   (dashboard upload only; never in the predict loop).
 - `src/config.py` — **deprecated** frozen v3 `Config` (different key names);
   do not add keys here.