⚡ Performance: Optimize vector allocation in filtered_const_roll_table

[perim]

💡 What: Added capacity reservation for the filtered and indices vectors inside the filtered_const_roll_table constructor, sizing them to mask.size().

🎯 Why: To avoid multiple intermediate memory allocations during push_back operations within the loop. This reduces memory fragmentation and saves CPU cycles that would otherwise be spent on reallocation and element copying.

📊 Measured Improvement:
I created a micro-benchmark instantiating filtered_const_roll_table 1,000 times with a vector/mask size of 100,000.

• Baseline: ~2903 ms
• Improvement: ~2867 ms
• Change over baseline: A savings of ~36 ms, which equates to roughly a 1.2% reduction in overall execution time for this heavy instantiation scenario.

---

PR created automatically by Jules for task 10831340825167954980 started by @perim

[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.