Summary

Adds a new Frontier-CS 2.0 vector_db_ann task for SIFT1M-scale approximate nearest-neighbor vector search, with a Rust project skeleton, hidden black-box evaluator, reference baseline, and Harbor support.

Also updates the Frontier-CS 2.0 Harbor adapter/environment so tasks can submit whole projects, configure runtime/judge dependencies, expose judge readiness, preserve iterative best submissions, and report task-specific metrics such as QPS, effective QPS, recall, load time, tokens, and cost.

Please read CONTRIBUTING.md before submitting.

Type of Change

• New research problem
• New algorithmic problem
• New Frontier-CS 2.0 problem
• Bug fix
• Documentation update
• Other: Harbor adapter/runtime support for project-based 2.0 tasks

Testing

• PYTHONPYCACHEPREFIX=/private/tmp/frontier-cs-pycache python3 -m py_compile ...
• uv run --no-sync frontier list 2.0
• uv run --no-sync frontier show 2.0 vector_db_ann
• Generated the Harbor task via frontier_cs_2_0.main --task-ids vector_db_ann
• Ran Harbor trials for vector_db_ann; best observed valid submission reached recall 0.9871, QPS 370.24, and score 20.65

Checklist

• Code follows the project structure and conventions
• Self-review completed
• Documentation updated (if applicable)

CI Validation (for new problems)

When adding new problems, CI will automatically validate that your reference solution achieves score > 0.

• Algorithmic problems: Include reference.cpp in your problem directory
• Research problems: Include reference.py (or reference.cpp if language: cpp in config.yaml)
• 2.0 problems: Include reference.py unless the problem config declares another language