v1.3.1

## [1.3.1] - 2025-09-12

### ✨ Features

**Full llama.cpp Support on All Platforms**
- Enabled complete llama.cpp support across Linux, Windows, macOS Intel, and macOS ARM64
- Resolved macOS ARM64 compilation issues with forked llama-cpp dependency
- Added ARM64-specific compiler capability detection and optimizations

**Enhanced Build System**
- Improved cross-platform compilation with specialized ARM64 handling
- Added comprehensive testing for macOS ARM64 llama.cpp integration
- Streamlined release workflow configuration for stable deployments

### 🐛 Bug Fixes

**macOS ARM64 Compilation Issues**
- Fixed GGML_ARM_I8MM compilation conflicts on Apple Silicon
- Resolved mixed-ISA build problems with targeted compiler flags
- Added proper target detection for ARM64 optimization features

**Release Workflow Stability**
- Enhanced release pipeline reliability across all supported platforms
- Fixed deployment configuration issues for stable v1.3.1 releases
- Improved error handling and fallback strategies

### 🔧 Technical Improvements

**Cross-Platform Compatibility**
- Updated llama-cpp dependency to specialized fork with ARM64 fixes
- Enhanced build.rs with platform-specific compilation logic
- Added comprehensive CMAKE configuration for different architectures

**Testing Infrastructure**
- Added isolated macOS ARM64 llama compilation testing
- Enhanced platform-specific build validation
- Improved error reporting for architecture-specific issues