⚡ Bolt: Optimize AdaptiveMemory with persistent connection

[Rohith-Shimori]

• 💡 What: Refactored AdaptiveMemory to use a persistent sqlite3 connection with WAL mode and thread locking, replacing the per-operation connection creation.
• 🎯 Why: Creating a new database connection for every operation is expensive and causes significant latency (~1.6ms per op).
• 📊 Impact: Reduces operation time by ~10x (from ~1.64ms to ~0.17ms per operation in benchmarks).
• 🔬 Measurement: Verified with benchmark_memory.py (300 operations: 0.49s -> 0.05s). Verified functionality with test_ananta_quick.py.

---

PR created automatically by Jules for task 14057141891828799364 started by @Rohith-Shimori

[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.