TRO-JSON v4: Slots je TRO und Staubereiche (jamAreas)

- model: Slot (separator/switch/scanner/drive + Rollen), JamArea (J<Separator>,
  Eingang eines TRO mit allen Vorgaengern), derive_slots, derive_jam_areas;
  Schema sps_skel-tro-v4, v3-Dateien werden weiter gelesen
- scl_context: ein JamEntr je eigenem Staubereich statt je Vorgaenger (wie
  UH01-UH05), Ausgaenge nennen den Staubereich des Nachfolgers
- Golden Files neu erzeugt, Tests fuer Slots/Staubereiche, Doku nachgezogen

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

CLAUDE.md

@@ -180,8 +180,9 @@ and status: `doc/SCL_Templates.md`. Golden files in `tests/fixtures/scl/` pin th
 byte — rewrite them (`SKEL_UPDATE_GOLDEN=1`) only for a reviewed, intended change.
 
 1. **`lib/tro_flow.py --json`** (implemented) — writes `<csv>_tro.json` (schema
-   `sps_skel-tro-v3`, `model.TroDocument`): TROs, system boundaries `SYSn`, connections with
-   their kind, findings. TRO-IDs come from `cfg/<csv>_tro_ids.ini` (`lib/tro_ids.py`): each
+   `sps_skel-tro-v4`, `model.TroDocument`): TROs with their hardware `slots`, system
+   boundaries `SYSn`, connections with their kind, `jamAreas` (one per TRO entry, key
+   `J<Separator>`, all predecessors that feed it), findings. TRO-IDs come from `cfg/<csv>_tro_ids.ini` (`lib/tro_ids.py`): each
    TRO is found again by the position of its separators, so IDs survive a new layout export
    (TeileIds do not); `--renumber` renumbers along the material flow, `--ids` points another
    export of the same plant at the same file. See `doc/Python_Scripts.md` §4.