fix: keep submission task scores strictly within range

Merge origin/main into training/unsloth (sync latest upstream)

feat: enhance TaskRunResponse with raw_total_reward and normalized score calculation

Merge origin/main into training/unsloth (safe local merge)

Merge origin/main into training/unsloth (safe local update)

fix: add missing build dependencies in Dockerfile for improved environment setup

feat: enhance Docker setup and environment configuration; add inference entrypoint and update YAML structure

fix: allow reset endpoint without request body

build: slim space dependencies and docker install

chore: refresh training outputs after local inference run

Update verification report: Fix training record count and environment validation status

Add training scripts and utilities for NodeAudit and GraphReview

Refactor inference and training modules; add Dockerfile and server setup

Add project planning document for GraphReview RL Environment with detailed specifications, phase plans, and requirements for hackathon submission.

feat: expand .env.example with comprehensive configuration settings for GraphReview environment

fix: accept HFTOKEN fallback for HF router auth

feat: auto-fallback inference to HF Router when HF_TOKEN is set

fix: allow inference training runs without local gguf weights in hosted env

fix: open selected module when clicking graph node

fix: improve graph tooltip formatting and report panel readability

fix: wrap long graph hover tooltips across report views

fix: make training bootstrap resilient in hosted runtime

fix: avoid training bootstrap 500 when local weights are unavailable in container

Refactor README.md: Replace old documentation with updated content for GraphReview environment, including project overview, architecture, tasks, and setup instructions. Remove obsolete README and defects4j references.

fix: add HF Spaces README front matter

fix: enable semgrep execution in local bin environments

feat: final submission — model fix, README, Dockerfile

feat: Implement validation for canonical fixture and add training suite

Add GraphReview report generation and semantic checks

Add Phase 06 Plan for Adaptive Judging and Edge Intelligence; Create initial project outline for GraphReview RL Environment

Add visualization and reporting features for Phase 05

feat: Update dependencies and add main entry point for server

feat: Enhance module resolution and grading configuration

feat: Implement task registry and grading system

feat: Implement chunking and graph management for code review environment

feat: initialize CodeReviewEnv with foundational components