Handle Resend rate limits, quotas, and error responses

Documented steps

1. Read rate-limit response headers on every call: ratelimit-limit, ratelimit-remaining, ratelimit-reset, retry-after
2. Default limit is 10 requests/second per team, shared across ALL the team's API keys
3. On 429 rate_limit_exceeded: wait retry-after seconds (it is expressed in SECONDS), then retry with backoff; queue requests client-side to stay under 10 rps
4. Distinguish 429 quota errors (daily/monthly sending quota exhausted — waiting seconds won't help) from rate-limit 429s
5. Use POST /emails/batch (100 emails = 1 request) to multiply effective throughput
6. Test failure handling with bounced@resend.dev and complained@resend.dev without hurting your domain reputation
7. Docs: https://resend.com/docs/api-reference/rate-limit and https://resend.com/docs/api-reference/errors