perf: reduce unnecessary daemon cloning in loops

[jdx]

Summary

• Optimize daemon collection iteration to reduce memory allocations

Problem

The check_retry and check_cron_schedules functions were cloning entire daemon collections upfront, even when only a small subset of daemons needed processing. For systems with many daemons, this caused unnecessary memory allocations.

Solution

Change the pattern from:

let daemons = state_file.daemons.clone(); // Clones ALL daemons
for (id, daemon) in daemons { ... }

To:

let ids: Vec<String> = state_file.daemons.iter()
    .filter(|...|) // Filter to only relevant daemons
    .map(|(id, _)| id.clone()) // Clone only IDs
    .collect();
for id in ids {
    let daemon = state_file.daemons.get(&id).clone(); // Clone only when needed
}

This reduces allocations when:

• Most daemons don't need retrying (only errored daemons with retries remaining)
• Most daemons don't have cron schedules

Test plan

• cargo build passes
• Manual test with many daemons

🤖 Generated with Claude Code

---

Note

Optimizes supervisor loops to minimize allocations and unnecessary cloning while preserving behavior.

• Refactors check_retry to collect only daemon IDs needing retries, then re-validate and clone on demand before invoking run
• Refactors check_cron_schedules to collect only IDs with cron settings, then fetch each daemon lazily for schedule evaluation and triggering
• Updates Cargo.lock (adds libc dependency entry)

^{Written by Cursor Bugbot for commit a9c2fa9. This will update automatically on new commits. Configure here.}

[cursor]

Cursor Bugbot has reviewed your changes and found 1 potential issue.

^{Bugbot Autofix is OFF. To automatically fix reported issues with Cloud Agents, enable Autofix in the Cursor dashboard.}