Cache and restore wallets

Improve login performance by caching wallet state on a per account level in unencrypted JSON file and rehydrate before wallets load.

Author: paullinator

docs/new-core-dbs-plan.md

@@ -0,0 +1,124 @@
+---
+name: Clean Up DB Requirements Doc
+overview: Transform the raw prompt in `new-core-dbs.md` into a structured requirements document with clear sections for background, requirements, constraints, and a placeholder for research results.
+todos:
+  - id: rewrite-doc
+    content: Rewrite new-core-dbs.md with proper requirements document structure
+    status: completed
+isProject: false
+---
+
+# New Core Database Requirements Document Cleanup
+
+## Current State
+
+The file [`docs/new-core-dbs.md`](docs/new-core-dbs.md) contains a raw research prompt with the following key information embedded:
+
+- **Target**: edge-core-js library
+- **Current storage**: disklet library (bridges to native on mobile, localStorage on web, filesystem on Node)
+- **Environments**: iOS/Android webview, Node.js, web browser
+- **Goal**: Add SQL or NoSQL database that's fast and efficient
+
+## Proposed Structure
+
+Rewrite as a proper requirements document:
+
+```markdown
+# Edge Core JS Database Requirements
+
+## Background
+
+Brief context about edge-core-js and why a database is needed.
+
+## Current Architecture
+
+- Disklet library overview
+- Platform bridging details
+
+## Requirements
+
+### Platform Support
+- iOS (headless WebView)
+- Android (headless WebView)
+- Node.js
+- Web browser
+
+### Performance
+- Fast read/write operations
+- Efficient storage
+
+### API Style
+- SQL or NoSQL (flexible)
+
+## Constraints
+
+- Must work within WebView JavaScript context on mobile
+- Must integrate with existing disklet architecture or replace it
+
+---
+
+## Research Results
+
+(To be completed)
+
+For each database option, include:
+- Platform compatibility details
+- Query capabilities (what types of queries are supported)
+- Performance characteristics
+- Integration complexity
+```
+
+## Changes
+
+1. Add proper markdown heading
+2. Extract background context into its own section
+3. Break out platform requirements explicitly
+4. Add constraints section based on the architectural details mentioned
+5. Add empty "Research Results" section at the end for future use
+
+---
+
+## Execution Log
+
+**Completed**: Rewrote `new-core-dbs.md` as a structured requirements document.
+
+**Changes made**:
+- Added proper title: "Edge Core JS Database Requirements"
+- Created Background section explaining edge-core-js purpose and need for database
+- Created Current Architecture section detailing disklet library and platform bridging
+- Created Requirements section with subsections:
+  - Platform Support (iOS/Android WebView, Node.js, Web browser)
+  - Performance (fast read/write, efficient storage, concurrent access)
+  - API Style (SQL/NoSQL flexible, JS/TS API, Promise-based)
+  - Query Capabilities (key-value, range queries, filtering, indexing)
+- Created Constraints section (WebView compatibility, disklet integration, offline-first, persistence)
+- Added Research Results placeholder with template for documenting each option
+
+---
+
+**Completed**: Added research results for 5 database options.
+
+**Databases Researched**:
+1. **Dexie.js** - IndexedDB wrapper, excellent cross-platform, NoSQL queries
+2. **sql.js** - SQLite via WebAssembly, full SQL support, manual persistence
+3. **PouchDB** - Document store with CouchDB sync, largest community
+4. **RxDB** - Reactive NoSQL, multiple backends, higher complexity
+5. **LokiJS** - In-memory database, fast but limited maintenance
+
+**Top Recommendations**:
+- Dexie.js for simplicity and balance
+- sql.js if full SQL is required
+- PouchDB if sync capabilities are needed later
+
+---
+
+**Completed**: Added Cordova WebView Migration Proposal.
+
+**Proposal Summary**:
+- Analyzed current architecture (custom WKWebView/WebView with bespoke message handlers)
+- Documented current native methods: disklet (6 methods), fetch, randomBytes, scrypt
+- Proposed migration to Cordova WebView for plugin ecosystem access
+- Outlined 3-phase migration: Cordova integration → SQLite integration → Testing
+- Identified challenges: RN+Cordova bridge, scrypt porting, migration complexity
+- Added alternative "Custom SQLite Bridge" approach as lighter option
+- Recommendation: Custom SQLite Bridge for database-only needs; full Cordova for broader plugin access