fix: resolve all typechecking errors across codebase

AGENTS.md

@@ -1,46 +1,26 @@
 # Tschausepp - Jass Card Tracker
 
-## Project Overview
-Mobile-first React web app for tracking Schaffhauser (Jass) card game rounds using camera and feature matching to detect and assign cards to players.
+Mobile-first React app for tracking Schaffhauser (Jass) card game rounds via camera.
 
 ## Tech Stack
-- React 18 (lightweight runtime)
-- TensorFlow.js (lite mode, CDN-hosted, ~5MB)
-- HTML5 Camera API (Mobile-friendly)
-- localStorage (JSON format, max 100 games)
+- React 18, TypeScript, Vite
+- TensorFlow.js (lite mode, CDN-hosted)
+- HTML5 Camera API (environment-facing preference)
+- localStorage (JSON, max 100 games)
 
-## Key Implementation Details
+## Developer Commands
+- `npm run typecheck`: Run TypeScript check
+- `npm run build`: Build production version
+- `npm run preview`: Preview production build
+- **Dev Server**: Always running on `localhost:5173`. Do not start manually.
 
-### Architecture
-- Component-based structure with 5 main screens:
-  1. Setup Screen - Player management and game configuration  
-  2. Camera Screen - Live preview with card detection overlay
-  3. Detection Component - Card boundary detection using image processing
-  4. Assignment Component - Radial sector layout with auto-assignment
-  5. Results Screen - Score calculation and game results
-  6. History Screen - Game storage and export functionality
+## Architecture & Logic
+- **Screens**: Setup → Camera (with Detection/Assignment) → Results → History.
+- **Detection**: Custom image processing + color analysis for suits (Schellen, Schilten, Eicheln, Rosen).
+- **Assignment**: Uses radial sectors to assign detected cards to the nearest player.
+- **Storage**: Max 100 games in localStorage; follows `games → rounds → cards` structure.
 
-### Development Commands
-- `npm run build` - Build production version
-- `npm run preview` - Preview production build
-- `npm run typecheck` - Run TypeScript type checking
-
-### Development Server
-- A development server is always guaranteed to be running on localhost:5173
-- **Do not start the development server manually under any circumstances**
-
-### Camera & Detection
-- Uses HTML5 Camera API with environment-facing camera preference
-- Card detection implemented with custom image processing techniques
-- Suit identification (Schellen, Schilten, Eicheln, Rosen) via color analysis
-- Auto-assignment based on radial sector + nearest player calculation
-
-### Data Storage
-- Games stored in localStorage with maximum of 100 games
-- Data structure follows: games → rounds → cards with player assignments
-- JSON format for export functionality
-
-### Mobile-First Design
-- Responsive UI optimized for mobile devices
-- Touch-friendly controls and interfaces
-- Camera access optimized for mobile environment
+## Reference Docs
+- `ML_SETUP_GUIDE.md`: ML pipeline architecture, training, and TF.js deployment.
+- `DETECTION_IMPROVEMENT_PLAN.md`: Roadmap for improving detection stability and accuracy.
+- `swiss_jass_suits.md`: Domain reference for Jass suit colors and iconography.