feat(tests): make inference_recorder into api_recorder (include tool_invoke) (#3403)

Renames `inference_recorder.py` to `api_recorder.py` and extends it to
support recording/replaying tool invocations in addition to inference
calls.

This allows us to record web-search, etc. tool calls and thereafter
apply recordings for `tests/integration/responses`

## Test Plan

```
export OPENAI_API_KEY=...
export TAVILY_SEARCH_API_KEY=...

./scripts/integration-tests.sh --stack-config ci-tests \
   --suite responses --inference-mode record-if-missing
```
This commit is contained in:
Ashwin Bharambe 2025-10-09 14:27:51 -07:00 committed by GitHub
parent 26fd5dbd34
commit f50ce11a3b
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
284 changed files with 296191 additions and 631 deletions

View file

@ -30,7 +30,7 @@ Options:
--stack-config STRING Stack configuration to use (required)
--suite STRING Test suite to run (default: 'base')
--setup STRING Test setup (models, env) to use (e.g., 'ollama', 'ollama-vision', 'gpt', 'vllm')
--inference-mode STRING Inference mode: record or replay (default: replay)
--inference-mode STRING Inference mode: replay, record-if-missing or record (default: replay)
--subdirs STRING Comma-separated list of test subdirectories to run (overrides suite)
--pattern STRING Regex pattern to pass to pytest -k
--collect-only Collect tests only without running them (skips server startup)