fix(control-station): separate lab contours from recordings

docs/09_OBSERVATION_SESSIONS.md

@@ -24,10 +24,12 @@ fully usable with only their available modalities.
 3. Stop acquisition normally, or allow the local service to recover an
    unexpected interruption on its next start. Session recording is automatic;
    the disk action is not required.
-4. Open **Данные → Сессии и записи**. The inline archive shows up to 100
-   indexed runs, their date, duration, state, modalities and background
-   preparation state. Opening an archive never replaces the live source owned
-   by **Наблюдение**.
+4. Open **Данные → Сессии и записи**. The compact **Сохранённые сессии**
+   dropdown shows up to 100 indexed runs, their date, duration, state,
+   modalities and background preparation state. The page itself remains one
+   full-size spatial viewer; the catalog is not expanded into an inline list.
+   Opening an archive never replaces the live source owned by
+   **Наблюдение**.
 5. Choose a replayable run. Opening a replay never performs conversion in the
    request. A ready recording opens immediately; otherwise the client receives
    HTTP 202, keeps the current scene mounted and polls the preparation status.
@@ -130,8 +132,10 @@ on deletion.
 The visual result is resolved by the LAB session id, so `LAB E19`, `LAB E21`
 and the derived temporal comparison `LAB E22` can coexist over `RAVNOVES00`
 without “latest accepted result” replacing an earlier experiment. The operator
-can open any row in **Данные → Сессии и записи** or the corresponding LAB row
-in **Тестировочный контур** and use the same
+can open any row in **Данные → Сессии и записи** or choose the corresponding
+published LAB work through **Тестировочный контур → Лабораторные контуры**.
+The test surface presents LAB works through `profile → work`, never as a copy
+of the saved-session list. A recorded LAB work then uses the same
 **Объекты 2D / Сегментация / Кубы 3D** controls. First publication must also
 warm the result-specific Rerun overlay; later opens reuse that verified cache
 and do not rerun AI inference.