chore(deps): update dependency com.github.haifengl:smile-kotlin to v6.3.0

[renovate]

This PR contains the following updates:

Package	Change	Age	Adoption	Passing	Confidence
com.github.haifengl:smile-kotlin (source)	6.1.0 → 6.3.0

---

Release Notes

haifengl/smile (com.github.haifengl:smile-kotlin)

v6.3.0: 6.3.0

Compare Source

✨ New Features

LLM

• SafeTensors Support: Added support for Hugging Face's SafeTensors format, including write functionality for efficient and secure tensor serialization.
• KV Caching: Memory-efficient key-value cache using a shared Radix-backed pool for faster LLM inference
• FlashAttention: Leverages FlashAttention for optimized attention computation

Serve - API

• Unified Model Catalog: All models (Chat, SMILE, ONNX) now listed via a single GET /api/v1/models endpoint with OpenAI-compatible format
• Conversation Management: Full CRUD operations for chat conversations via the new Conversation resource
• Streaming & Non-Streaming: Chat completions support both streaming (SSE) and non-streaming JSON responses

Serve - Inference UI

• Unified Inference Page: New /infer page combines Chat, SMILE, and ONNX inference in one interface
• Vision Mode: ONNX inference now supports image inputs with automatic preprocessing (resize, normalization)
• Virtualized Results: Large batch results display efficiently with virtual scrolling
• Split Layout: Form and results display side-by-side for better workflow

Docker & Deployment

• GPU Docker Image: Multi-stage build with CUDA 13.2 support and LibTorch
• CPU Docker Images: Upgrade to UBI 10 with OpenJDK 25 runtime

---

🔧 Improvements

Performance

• Faster File Sync: SafeTensors loading now uses positioned FileChannel I/O instead of memory-mapping (handles multi-GB checkpoints)
• Optimized Memory: ONNX Runtime memory footprint restricted; GPU memory fraction configurable

Developer Experience

• Better Error Messages: ONNX runtime loading failures now show clear setup hints
• Tokenizer Auto-Download: HuggingFace tokenizers downloaded automatically when missing
• Model Architecture Info: GET /api/v1/models/{id} now returns detailed model metadata (architecture, shapes, etc.)

---

🐛 Bug Fixes

Core

• Big-Endian Float Corruption: Fixed ONNX tensor byte order handling (was causing scrambled predictions on big-endian systems)
• SSE Race Conditions: Chat streaming now subscribes before generation starts, preventing lost chunks

Testing & Reliability

• Test Port Conflicts: Fixed flaky tests caused by port 8081 being in use (now picks random free port)
• Association Rules Tests: Fixed race conditions in FP-Growth tests causing flaky counts

HuggingFace Integration

• Broken Symlinks: Fixed tokenizer resolution when snapshot links were incorrectly computed
• Model Validation: Chat service now fails cleanly with 503 when model is unavailable

Studio

• Code Cells: add null safety and fix evalMagic in code cells by @​TsybulkaM in #​884
• Theme Setting: Fixed theme setting logic in SettingsDialog by @​TsybulkaM in #​885

---

🔒 Security

• Native Library Loading: Improved error handling for CUDA and ONNX native library initialization

---

⚠️ Breaking Changes

• Model List Endpoints: GET /api/v1/ml/models and GET /api/v1/onnx removed; use GET /api/v1/models instead
• Scala VegaLite API: Moved from smile.plot.vega to smile.vega (updated imports required)

---

📦 Dependencies

• Torch: Upgraded to 2.13
• ONNX Runtime: Upgraded to 1.29
• SBT: Upgraded to 2.0.6
• CUDA: Support for CUDA 13.2

---

Full Changelog: haifengl/smile@v6.2.5...v6.3.0

v6.2.5: 6.2.5

Compare Source

What's Changed

• Fix Smile Studio notebook cells accumulate a trailing blank line on save by @​TsybulkaM in #​873
• Fix macro and weighted F-score to average per-class scores by @​gaoflow in #​874
• Fix macro/weighted F1 and two sibling averaging bugs in deep.metric by @​gaoflow in #​877
• Fix tail p-values collapsing to 0 in Gamma/GLM, and NegativeBinomial pmf overflow by @​gaoflow in #​878
• Fix MatthewsCorrelation int overflow and zero-denominator NaN by @​gaoflow in #​879
• Fix int overflow in CorTest.spearman and CorTest.kendall for large samples by @​gaoflow in #​880
• Bump linkify-it from 5.0.1 to 5.0.2 in /website in the npm_and_yarn group across 1 directory by @​dependabot[bot] in #​875
• Bump the npm_and_yarn group across 1 directory with 2 updates by @​dependabot[bot] in #​876

Thanks @​gaoflow and @​TsybulkaM for great contributions!

Full Changelog: haifengl/smile@v6.2.4...v6.2.5

v6.2.4: 6.2.4

Compare Source

✨ New Java Coding Skills

• Java Migration: Assist with migrating Java codebases to newer versions
• Architecture Review: Automatically analyze code structure and design patterns for best practices
• Concurrency Review: Analyze multi-threaded code for race conditions and deadlocks
• JPA Patterns: Validate Java Persistence API implementations and suggest optimizations
• Logging Patterns: Review logging strategies and ensure proper observability
• Performance Audit: Identify performance bottlenecks and optimization opportunities
• Security Audit: Detect potential security vulnerabilities in your code
• Test Quality with AssertJ: Improve test assertions using AssertJ's fluent API

---

🔧 Improvements

• GPT-5.6 Model Family: Support for the latest GPT-5.6 model family in the LLM integration.
• Claude Sonnet 5: Added support for Claude Sonnet 5 model.

---

🐛 Bug Fixes

• Fixed UI Selection Issue: Resolved issue with Settings dialog.

---

New Contributors

• @​bglglzd made their first contribution in #​870

Full Changelog: haifengl/smile@v6.2.3...v6.2.4

v6.2.3: 6.2.3

Compare Source

✨ New Features

• Dark Theme: Default dark theme.
• Neighborhood Functions: Added square, triangle, and Mexican hat neighborhood functions.
• Spark Module: Published smile-spark module and migrated to Spark v4 API.
• BigQuery MCP Server: Added BigQuery MCP server integration.

🔧 Improvements

• Hadoop Compatibility: Updated to Hadoop 3.5 for Java 21/25 compatibility.
• Javadoc Clarification: Improved documentation for MemorySegment handles in Tensor constructor.
• UI Scaling: Disabled uiScale = 1 for better HiDPI scaling support.
• Toolbar Visibility: Toolbar no longer shows on Windows/Linux.

🛠 Miscellaneous

• Golden Numbers: Updated golden numbers.

Full Changelog: haifengl/smile@v6.2.2...v6.2.3

v6.2.2: 6.2.2

Compare Source

✨ New Features

• AWS Bedrock Support: Support Anthropic models hosted by AWS Bedrock.
• Public API Enhancements: Made SOM.Neuron and bmu() public for easier integration and customization.

🏷️ Other Changes

• Dependency Updates: Updated to torch 2.12.1 and onnx 1.26.0 for improved compatibility.
• Clarans Algorithm: Renamed clarans to kmedoids for clarity for Scala and Kotlin API.

Full Changelog: haifengl/smile@v6.2.1...v6.2.2

v6.2.1: 6.2.1

Compare Source

✨ New Features

• AWS Bedrock Bearer Token Support: Added support for AWS_BEARER_TOKEN_BEDROCK authentication, enabling seamless integration with AWS Bedrock models.
• Clojure Association Module: New association.clj module added to the Clojure API for association rule mining capabilities.

🔧 Improvements

• Frontier Models Updated: Latest frontier AI models are now available in the platform.
• Clojure API Updated to 6.2.1: Clojure bindings synchronized with the latest core API version.
• Notepad Font Size Synchronization: The built-in notepad now responds to system font size changes, improving accessibility.

🐛 Bug Fixes

• Enhanced Agent LLM API: Safe handling when no LLM is configured.

📦 Dependency Updates

• markdown-it: Updated to 14.2.0 (website)
• vite: Updated to 8.0.16 (serve/webui)

🔨 Maintenance

• Added Clojure unit tests for improved test coverage
• Updated CI/CD pipeline with npm install step
• Synchronized package.json and package-lock.json for consistent builds
• Documentation updates and cleanup

Full Changelog: haifengl/smile@v6.2.0...v6.2.1

v6.2.0: 6.2.0

Compare Source

✨ New Features

• Deep Learning Integration: Introduced the smile.torch module and binaries, enabling advanced deep learning capabilities. This is a lower binding module. The users should continue to use higher-level modules such as smile.deep, smile.llm, smile.vision, etc.
• Docker GPU Support: Added Dockerfile.jvm-gpu and matrix strategy for publishing GPU-enabled Docker images.

🔧 Improvements

• Memory Management: Enhanced memory handling and fixed memory leaks in deep learning modules.
• Tensor Management: Improved input tensor memory management and safer residual calculation across mixed backends.
• Dependency Updates: Upgraded Quarkus to 3.36.2 and Kotlin to 2.4.0.
• Build & Environment: Improved environment variable handling, permissions configuration, and test workflows.
• Examples: Added new deep learning example scripts (examples/deep.jsh).
• NoGradGuardTest: Added new test for gradient management.

🛡️ Security

• Improved permissions configuration for safer builds.

Full Changelog: haifengl/smile@v6.1.1...v6.2.0

v6.1.1: 6.1.1

Compare Source

✨ New Features

• Python Skills Expansion: 10+ new Python skills with code review, API design, security audit, performance optimization, etc.
• Find and Replace in SmileStudio: Added find and replace functionality for easier text editing
• File Operations in Context Menu: New options to create files and directories directly from the context menu
• Open Files with Default Application: Open files using your system's default application
• TotalSupportTree Class: was package visible. Changed to public accessible.

---

🔧 Improvements

• Internationalization: Enhance i18n support in SmileStudio, Workspace, and Notepad
• Multi-thread Safe Edit Distance: Make EditDistance multi-thread safe

---

🔄 Dependency Updates

• Updated various project dependencies for security improvements
• Quarkus 3.35.4: Upgraded to latest Quarkus for better performance and stability
• Apache Arrow 19.0.0: Updated to latest Arrow for improved performance

---

📦 Build & Infrastructure

• Gradle 9.5.1: Updated to the latest Gradle version for improved build performance
• JVM Options Standardization: Consistent JVM options across the project
• Code Coverage: Added JaCoCo plugin with report aggregation for better test coverage visibility
• Classification Documentation: Revised and improved classification documentation
• Disabled parallel builds to optimize memory usage for CodeQL analysis

Full Changelog: haifengl/smile@v6.1.0...v6.1.1

---

Configuration

📅 Schedule: (UTC)

• Branch creation
  • At any time (no schedule defined)
• Automerge
  • At any time (no schedule defined)

🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.

♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.

🔕 Ignore: Close this PR and you won't be reminded about this update again.

---

• If you want to rebase/retry this PR, check this box

---

This PR was generated by Mend Renovate. View the repository job log.