fix: detect Mercury stage per-view via session name

The Mercury app (3.2.x) is a hybrid server whose kernel pool serves both
the client-facing app view and JupyterLab, so the server-level
MERCURY_CONFIG_DIR signal cannot tell who is looking and leaks backstage
content in the app. Switch the primary stage signal to the shadow-copy
session name (__mercury__ in JPY_SESSION_NAME), keeping the env var only
as a --working-dir fallback.

- Update CX Discovery staging.py to the per-view detection model
- Ignore Mercury runtime artifacts (.mercury_sessions/, *__mercury__*)
- Document the leak in remaining masters (TEI, CTM, AI Diagnostic,
  template) and flag the staging.py retrofit as urgent

CLAUDE.md

@@ -78,9 +78,11 @@ there; the repo-root `.venv` (from `make setup`) serves only `core/` and the not
 structural suite. Running a master's pytest from the wrong venv is the classic
 "missing module" ghost.
 
-- `MERCURY_CONFIG_DIR` present in the environment = **the stage is live** (a client
-  may be looking). Its presence is the stage signal `staging.py` keys off — never set
-  it manually except to simulate the stage in a test.
+- Stage detection is **per view, not per server**: the primary signal is
+  `__mercury__` in `JPY_SESSION_NAME` (the app's shadow-copy session);
+  `MERCURY_CONFIG_DIR` is only a fallback and is set **only** by
+  `mercury --working-dir`. Never set either manually except to simulate the stage
+  in a test (Mercury pattern §6 has the mechanism).
 - Only masters live here. **If a notebook in this repo contains a real client's name,
   something is wrong** — stop and flag it (see Confidentiality).
 
@@ -219,6 +221,12 @@ quietly leaving it. Live ones worth knowing:
   fix the typing with the redesign, not piecemeal.
 - **TEI twins + CTM pre-date the tagged-cell taxonomy** — the structural suite
   grandfathers them by name in `tests/nbcheck.py`, each with its reason.
+- **Every master except CX Discovery still carries the env-var-only `staging.py`**
+  (TEI twins, CTM, AI Diagnostic, template) — their `on_stage()` misses the Mercury
+  app unless the server was launched with `--working-dir`, so **their stages leak
+  backstage content** on a plain `mercury` launch (found live 2026-07-31; fixed in
+  the Discovery reference). The retrofit is a recorded opportunity — until it
+  lands, serve those masters only via `mercury --working-dir .`.
 - **`template/MercuryNotebook/` encodes only the py-engine model** (and its
   `staging.py` lacks the mypy-strict `backstage_md` variant); rework is a recorded
   follow-up. Its `exports/*.{html,md}` are tracked — predates the exports rule.
@@ -237,7 +245,9 @@ quietly leaving it. Live ones worth knowing:
 - TEI + AI Diagnostic redesigns to the notebook-first model (each moves its notebook
   from GRANDFATHERED to NOTEBOOK_FIRST in `tests/nbcheck.py`).
 - Template rework: an Assessment template derived from the Discovery reference;
-  staging.py mypy-strict retrofit across masters.
+  staging.py retrofit across masters (now urgent-ish: brings the corrected
+  shadow-session stage detection, plus mypy-strict + `backstage_md`, to the TEI
+  twins, CTM, AI Diagnostic, and the template).
 
 ## Reference