Make benchmark link count configurable: 10 for PRs, 1000 for main branch

[konard]

📋 Summary

This PR implements a configurable benchmark link count to optimize CI/CD performance:

• Pull Requests: Execute benchmarks with 10 links per operation for faster feedback
• Main/Master Branch: Execute benchmarks with 1000 links per operation for comprehensive performance data

🔧 Implementation Details

Changes Made

1. Modified rust/src/lib.rs:

   • Changed LINK_COUNT from a constant (const) to a lazy-initialized static that reads from the BENCHMARK_LINK_COUNT environment variable
   • Defaults to 10 links when the environment variable is not set
   • Uses once_cell::Lazy for thread-safe initialization

2. Updated benchmark files:

   • Modified all benchmark code to dereference the lazy static using *LINK_COUNT
   • Files updated: create.rs, delete.rs, update.rs

3. Modified .github/workflows/rust.yml:

   • Added BENCHMARK_LINK_COUNT environment variable to the "Run benchmark" step
   • Conditionally sets value based on GitHub context:
     • 1000 when pushing to main or master branch
     • 10 for all pull requests

4. Code formatting:

   • Applied cargo fmt to ensure all Rust code follows the project's formatting standards

Benefits

• Faster PR feedback: Benchmarks run ~100x faster in PRs (10 links vs 1000 links)
• Accurate main branch data: Full 1000-link benchmarks preserved for merged code
• Configurable: Can be overridden locally via environment variable for testing

Testing

• ✅ Code compiles successfully (cargo check)
• ✅ Formatting applied (cargo fmt)
• ✅ Changes verified to match issue requirements

📚 Technical Approach

The implementation uses Rust's Lazy<T> type from the once_cell crate to initialize the link count once at runtime by reading the environment variable. This approach:

• Maintains backward compatibility (defaults to 10)
• Allows runtime configuration without recompilation
• Thread-safe initialization
• Zero runtime overhead after first access

🔗 Related

Fixes #11

---

🤖 Generated with Claude Code

Co-Authored-By: Claude noreply@anthropic.com

[konard]

🤖 Solution Draft Log

This log file contains the complete execution trace of the AI solution draft process.

💰 Cost estimation:

• Public pricing estimate: $1.497674 USD
• Calculated by Anthropic: $0.889932 USD
• Difference: $-0.607742 (-40.58%)
  📎 Log file uploaded as GitHub Gist (382KB)
  🔗 View complete solution draft log

---

Now working session is ended, feel free to review and add any feedback on the solution draft.