Add rate limiting and cost guardrails for Bedrock AI calls

[llinsss]

Problem

/api/v1/hint and /api/v1/story both default use_bedrock=True and have zero rate limiting. A misbehaving or retry-happy frontend (or a malicious caller, given auth isn't in place yet) can trigger unbounded Bedrock invocations with no cap, risking a surprise AWS bill.

Scope

• Add per-student and/or per-API-key rate limiting on Bedrock-backed endpoints (e.g. token bucket, N calls per minute).
• Add a global daily/monthly Bedrock call budget with a hard cutover to template-only fallback once exceeded (log when this happens).
• Expose current usage via an admin endpoint (e.g. GET /api/v1/admin/bedrock-usage).
• Make limits configurable via environment variables, with sane documented defaults.

Acceptance criteria

• Exceeding the per-student rate limit returns a 429 (or falls back to template mode, your call — document the choice).
• Exceeding the global budget forces template-only behavior for all requests until the window resets (test simulates budget exhaustion).
• Usage counters are observable via the admin endpoint.

[Xaxxoo]

Hi, Can I take up on this issue ?

[OAKVISUALZ]

Please assign me this issue.

[odiliflash-ops]

I would love to work on this issue

[grantfox-oss]

🦊 GrantFox — @odiliflash-ops has been assigned to this issue as part of the Third Campaign campaign!

Next steps:

1. Open a Pull Request referencing this issue (e.g., Closes #7)
2. Your PR will be reviewed by the DogStark maintainers

Good luck! Track your progress on GrantFox.

[grantfox-oss]

🎉 This issue has been marked as completed on GrantFox!

@odiliflash-ops's PR #41 was approved and merged by @llinsss.

🏆 @odiliflash-ops: You earned 35 FoxPoints for this contribution! Your current tier: Explorer (376 total points). Track your full progress on GrantFox.

👏 Great work, @odiliflash-ops! Keep contributing to DogStark.