fix: bump rate limits in reqresp batch retry test to prevent flaky failures

[AztecBot]

Summary

PR #21322 mocked the failing requests (indices 10, 11) to deterministically return RATE_LIMIT_EXCEEDED, but the passing requests (indices 0-9) still go through originalSend to real peers. With 10 requests split across only 2 peers (~5 each), the responding nodes' GCRA rate limiter (default: 1 per 200ms) kicks in and rate-limits some legitimate requests, causing non-deterministic test failures.

Fix: add high rate limit quotas (50 per 1000ms) to the test nodes — the same pattern the adjacent "pinned peer" test already uses.

Test plan

• Same test file, same pattern as the adjacent should send a batch request with a pinned peer test which already passes with these rate limits
• Could not build locally (no Docker daemon for bb-cpp-wasm), relying on CI to validate

ClaudeBox log: https://claudebox.work/s/ae01953d584e6acc?run=1