Don't retry 429s returned from the cache service

[GhadimiR]

These are only returned in exceptional situations, it's better to not slow down the run retrying these when cache is best effort.

[Copilot]

Pull request overview

This PR modifies the cache service client to stop retrying 429 (Too Many Requests) responses from the cache service. Previously, 429s were treated as retryable errors, but this change makes them fail immediately to prevent slowing down runs since cache is a best-effort service.

Changes:

• Removed HttpCodes.TooManyRequests from the list of retryable status codes
• Added special handling for 429 responses to warn users about rate limits and throw immediately
• Added comprehensive test coverage for the new 429 handling behavior

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 2 comments.

File	Description
packages/cache/src/internal/shared/cacheTwirpClient.ts	Removed 429 from retryable status codes and added special handling to detect retry-after headers and warn users before throwing
packages/cache/tests/cacheTwirpClient.test.ts	Added comprehensive tests validating that 429s fail immediately without retries and properly handle retry-after headers
packages/cache/package.json	Bumped version from 5.0.2 to 5.0.3
packages/cache/RELEASES.md	Added release notes for 5.0.3

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Link-]

For posterity, we're not retrying on the retry after schedule to prevent workflow run delays. It's best to fail fast in this case. We might want to make this configurable though.