⚡ Bolt: Optimize candidate filtering by removing unnecessary clones

[google-labs-jules]

💡 What: Refactored the Filter trait and its implementations to return original candidates upon error, eliminating the need for preemptive cloning in the pipeline loop.
🎯 Why: candidates.clone() was being called for every filter in the pipeline as a recovery mechanism. Since candidates are large vectors of complex structs, this was a significant performance bottleneck.
📊 Impact: Eliminates N clones per request where N is the number of filters (currently ~12). This should significantly reduce CPU usage and memory allocations.
🔬 Measurement: Verify that the pipeline still functions correctly and handles errors (if any) by restoring candidates.

---

PR created automatically by Jules for task 3257336148489922122 started by @sashimikun

[google-labs-jules]

👋 Jules, reporting for duty! I'm here to lend a hand with this pull request.

When you start a review, I'll add a 👀 emoji to each comment to let you know I've read it. I'll focus on feedback directed at me and will do my best to stay out of conversations between you and other bots or reviewers to keep the noise down.

I'll push a commit with your requested changes shortly after. Please note there might be a delay between these steps, but rest assured I'm on the job!

For more direct control, you can switch me to Reactive Mode. When this mode is on, I will only act on comments where you specifically mention me with @jules. You can find this option in the Pull Request section of your global Jules UI settings. You can always switch back!

New to Jules? Learn more at jules.google/docs.

---

For security, I will only act on instructions from the user who triggered this task.

[sashimikun]

@jules write me a before and after test

[google-labs-jules]

@jules write me a before and after test

I've added a test case test_filter_error_recovery in candidate-pipeline/candidate_pipeline.rs. This test demonstrates that the pipeline correctly recovers the original candidates when a filter fails, without needing the preemptive candidates.clone() that was previously required for every filter execution. While I can't easily reproduce the 'before' state performance (since I've updated the trait signature), this test confirms that the new 'return ownership on error' pattern works as intended.