feat(alerts): run the llm detector on claude haiku 5.5

feat(alerts): run the llm detector on claude haiku 5.5 ## Problem - Every check by an alert on the AI detector makes one billed model call, so an hourly AI alert costs every hour. - The detector runs on Claude Sonnet 5, at $2 / $10 per million input / output tokens. - Claude Haiku 5.5 costs $0.10 / $0.50 per million tokens for prompts up to 100K tokens ✨ Submitted to Merge by @andrewm4894. It will be added to the merge queue once all branch protection rules pass. See more details [here](https://app.trunk.io/posthog-inc/merge-queue/3921a8a3-abf7-42ff-b9cf-ef4fab8f3649/115139). ### PostHog Review · Standard · reviewed `0e632a9` Nothing to post. | Step | Status | Result | |---|---|---| | Prepare the diff | done | 1 file | | Main review and lenses | done | 0 issues | | Merge and cap | done | 0 → 0 | | Publish | done | Nothing to post | ## 🤖 CI report ⚠️ Trunk lane — backend Python lane This PR is assigned to the backend Python lane. It runs backend Python tests ...