Enforce response latency and reuse timings in later tests.
Fail a test if it exceeds the threshold (ms):
```toml
max_response_time = 200
```
After each test, Catalyst sets `response_time_ms` which can be used in subsequent requests or logs:
```toml
body = { "prev_ms" = "{{response_time_ms}}" }
```