feat(saas): multi-tenant isolation (Phase 1) + rate-limit/audit/org-scoped export (Phase 2)

Phase 1 (tenant isolation):
- g.org_id set on require_auth; assert_tenant()/current_org_id() choke-point helpers.
- Multi-org provisioning: POST /api/admin/users {new_org:true} (super_admin) creates a
  new org + its first admin; GET /api/admin/orgs (super_admin sees all, admin own).
- Fixed latent create_org double-id bug (dict id != store key).
- test_saas_tenant.py: org2 admin blocked from org1 group (403), can't list org1
  groups/users, sees only own org; super_admin sees all.

Phase 2 (hardening):
- Rate limit login (per-IP + per-username) + chat send (per-user) to protect LLM cost
  and slow brute force; services/rate_limit.py (in-memory + disk, no external deps).
- Audit log data/audit/audit.jsonl on org.create, user.promote_super_admin, analytics.export.
- CSV export now org-scoped (admin exports only own org).
All 8 backend suites pass.

backend/app/api/auth_routes.py

@@ -27,6 +27,14 @@ def login():
     password = data.get("password") or ""
     if not username or not password:
         raise ApiError("username and password are required")
+    # Slow down brute-force / abuse: per-IP + per-username window.
+    from ..services.rate_limit import check as ratelimit
+
+    client_ip = request.remote_addr or "?"
+    if not ratelimit("login:ip", client_ip, limit=15, window=300):
+        raise ApiError("too many attempts, try again later", 429)
+    if not ratelimit("login:user", username, limit=8, window=300):
+        raise ApiError("too many attempts, try again later", 429)
     try:
         user = _store().verify(username, password)
         token = _store().issue_token(user)