Use IDB for query storage (#9687)

* Add storage abstraction for persisted query data

Introduce a platform-specific storage abstraction layer for react-query
persistence:
- Native: Uses MMKV for high-performance synchronous storage
- Web: Uses IndexedDB via the `idb` library for efficient async storage

This replaces the previous AsyncStorage implementation with more performant
platform-native solutions. The abstraction maintains API compatibility with
@tanstack/query-async-storage-persister.

* Refactor storage abstraction to use factory pattern

Change createPersistedQueryStorage to a factory function that accepts a
storage ID, allowing multiple isolated storage instances:
- Native: Each instance gets its own MMKV store
- Web: Each instance gets its own IndexedDB database

Adopt the factory pattern in:
- react-query.tsx: Uses 'persisted_queries' storage
- ageAssurance/data.tsx: Uses 'age_assurance' storage

This provides better separation between different query client caches
and allows each to be managed independently.

---------

Co-authored-by: Claude <noreply@anthropic.com>

package.json

@@ -173,6 +173,7 @@
     "history": "^5.3.0",
     "hls.js": "^1.6.2",
     "idb-keyval": "^6.2.2",
+    "idb": "^8.0.3",
     "js-sha256": "^0.9.0",
     "jwt-decode": "^4.0.0",
     "lande": "^1.0.10",