Phase 4 checkpoint: chat-screen perf refactor + retryable blast-failure + repo-wide dispose-ref guardrail

Chat-screen performance (customer + mitra):
- Parent screens have zero `ref.watch` — only `ref.listen` for side effects
- Body extracted into its own `ConsumerStatefulWidget`; AppBar parts split
  into narrow `.select` consumers (mode, sensitivity, timer)
- Per-second timer ticks routed to dedicated providers
  (`chatRemainingSecondsProvider` + new `mitraChatRemainingSecondsProvider`)
  so WS `session_tick` frames don't invalidate the rest of the chat state

Dispose-in-ref bug fix:
- `home_screen.dart`, `payment_screen.dart`, `mitra_chat_screen.dart` —
  ref-using cleanup moved from `dispose()` to `deactivate()`. Modern
  Riverpod invalidates `ref` the moment `dispose()` runs; the resulting
  silent error corrupts the widget-tree finalize and the next screen
  appears frozen
- `halo_lints` package added at repo root with `no_ref_in_dispose` rule
  to catch this pattern in CI / IDE analysis
- `custom_lint` activated in both apps' `analysis_options.yaml`
  (was installed but never wired in — also brings `riverpod_lint`'s
  `avoid_ref_inside_state_dispose` online)
- CLAUDE.md Pitfalls section added to client_app + mitra_app

Phase 4 §3 retryable blast-failure (Option A):
- Backend `expirePairingRequest` + all-rejected use
  `recordIntermediateFailure` instead of `failPaymentSession` so the
  payment session stays `confirmed` for re-blast
- WS `pairing_failed` payload carries `is_terminal: false` on the
  retryable paths; client parses the flag and exposes `retryBlast()`
- "Coba cari lagi" CTA on S7 Timeout now re-blasts on the same payment
- Pairing service test updated to reflect the new semantics

Customer waiting-payment screen navigation patch:
- `_navigateTerminal` uses `Future.microtask` + `addPostFrameCallback`
  redundancy after a release-mode bug where polling stopped but
  `context.go` never fired, leaving the screen visually stuck on
  "menunggu pembayaran"

See requirement/resume-2026-05-15.md for next-day pickup checklist
(mitra release rebuild + S21 Ultra install + retest is the gating item).

Bundles unrelated in-flight Phase 4 §2.x work that was already on disk
(ESP screen removal, USP one-time gate scaffolding, bestie-availability
public route, OTP service edits, Maestro flow tweaks) — kept together
to avoid a partial-rebase mess.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

backend/src/routes/public/client.mitra-availability.routes.js

@@ -3,15 +3,17 @@ import { countAvailableMitrasFromCache } from '../../services/mitra-status.servi
 import { UserType } from '../../constants.js'
 
 /**
- * Customer-home availability poll.
+ * Customer-authed availability poll (kept for CC/debug callers that want the
+ * raw count).
  *
- *   GET /api/client/mitra-availability  →  200 { available: bool, count?: number }
+ *   GET /api/client/mitra-availability  →  200 { available: bool, count: number }
  *
- * Hot endpoint by design — polled every 5s per active customer while their home is
- * foregrounded. Backed by a 10s in-memory cache (see mitra-status.service.js) so DB load
+ * The customer home polls `/api/public/bestie/available` instead — that route
+ * is unauthenticated and returns only the boolean, since SHome1st renders
+ * before the user has any JWT (see `requirement/flow_customer.mermaid.md` §1).
+ *
+ * Backed by a 10s in-memory cache (see mitra-status.service.js) so DB load
  * stays bounded regardless of poller count. No rate limit by intent.
- *
- * `count` is included for CC/debug; the customer UI must read only `available`.
  */
 export const clientMitraAvailabilityRoutes = async (app) => {
   app.get('/', { preHandler: [authenticate] }, async (request, reply) => {

backend/src/routes/public/public.bestie-availability.routes.js

@@ -0,0 +1,25 @@
+import { countAvailableMitrasFromCache } from '../../services/mitra-status.service.js'
+
+/**
+ * Public bestie-availability beacon.
+ *
+ *   GET /api/public/bestie/available  →  200 { available: bool }
+ *
+ * Unauthenticated by design: the SHome1st CTA must reflect global availability
+ * BEFORE the user has any JWT (see `requirement/flow_customer.mermaid.md` §1 +
+ * router.dart's "fresh / unauthenticated users land on Home directly" carve-out).
+ *
+ * Output is intentionally a single boolean — no `count`, no IDs, no metadata —
+ * so this endpoint leaks no operational signal beyond "at least one bestie is
+ * online right now". Backed by the same 10s in-memory cache that bounds DB
+ * load regardless of poller count.
+ *
+ * The auth'd `/api/client/mitra-availability` route is kept for CC/debug
+ * callers that need the raw count.
+ */
+export const publicBestieAvailabilityRoutes = async (app) => {
+  app.get('/available', async (_request, reply) => {
+    const { available } = await countAvailableMitrasFromCache()
+    return reply.send({ success: true, data: { available } })
+  })
+}