Stabilize OpenAI-compatible streamed tool-use test

[myzie]

Summary

• force the named add tool in the streamed tool-use integration test
• forward parallel_tool_calls through the OpenAI-compatible request adapter and disable parallel calls for this single-call test
• add request-configuration regression coverage
• update all legacy assert.Equal calls in openai_test.go to Wonton v0.2.1 actual-first ordering

Why

The live test used automatic tool choice while requiring exactly one tool call. OpenAI can validly answer without a tool in that mode, which caused the main-branch CI failure. The assertion arguments were also reversed, so the reported want and got values were misleading.

Validation

• go test ./providers/openaicompletions -count=1
• go test ./providers/openaicompletions -run TestToolUseStream -count=10
• go test ./... -count=1
• go vet ./...

Summary by CodeRabbit

• New Features

  • Tool-enabled requests can now explicitly control whether multiple tool calls may run in parallel.
  • Added support for the optional parallel_tool_calls request setting.

• Tests

  • Expanded coverage for forced single tool calls, prompt caching, streaming tool calls, message conversion, reasoning replay, and usage-token reporting.
  • Updated test assertions for consistency.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Essentials

Run ID: 7aeed014-2a31-487b-937e-f6c78034ed4e

📥 Commits

Reviewing files that changed from the base of the PR and between 1e08ec9 and 7617fe6.

📒 Files selected for processing (3)

• providers/openaicompletions/openai.go
• providers/openaicompletions/openai_test.go
• providers/openaicompletions/types.go

Included review availability: Your plan provides up to 5 included reviews per hour; 0 remain after this review.

---

📝 Walkthrough

Walkthrough

The OpenAI completions provider now propagates the configured parallel tool-call setting. The request type supports the corresponding JSON field. Tests add forced single-tool coverage and normalize assertion ordering.

Changes

OpenAI completions tool-call handling

Layer / File(s)	Summary
Parallel tool-call request wiring providers/openaicompletions/types.go, providers/openaicompletions/openai.go, providers/openaicompletions/openai_test.go	Request now supports parallel_tool_calls. Tool-enabled requests copy the configured value. Tests cover explicit single-tool selection and streaming responses.
Completions test validation providers/openaicompletions/openai_test.go	Tests use the preferred assertion argument order and validate tool conversion, reasoning replay, usage tokens, cache counts, and streaming behavior.

Estimated code review effort: 2 (Simple) | ~10 minutes

Merge Risk: ⚪ Minimal · up to 7617f

The OpenAI-compatible adapter now carries the parallel tool-call setting, while streamed single-tool tests explicitly select the tool and disable parallel calls. No current merge-blocking risk is identified.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 17.24% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 29 functions across 3 files.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly and concisely describes the primary change: stabilizing the OpenAI-compatible streamed tool-use test.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

• Fix all pre-merge checks with AI

✨ Finishing Touches 💡 1

📝 Generate docstrings 💡

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch codex/reliable-tool-use-stream-test

---

_{Comment @coderabbitai help to get the list of available commands.}