test(e2e): stabilize Mistral required streaming

[CatherineSue]

Description

Problem

After PR #1478 merged, TestToolChoiceMistral.test_tool_choice_required_streaming started showing flakiness: required streaming sometimes produced no streamed tool-call chunks. The comparable non-streaming required test already pins temperature=0.2, and the stricter required streaming arguments test pins temperature=0.1; this smoke test was still using the backend default sampling temperature.

Solution

Pin the Mistral required streaming smoke test to temperature=0.1 so the assertion exercises streaming tool-call plumbing instead of sampling variance. I also checked the Mistral e2e setup: it uses the model default chat template and --tool-call-parser mistral; no special chat template is injected. The Mistral path still uses the structural-tag constraint from MistralParser::build_structural_tag.

Changes

• Set temperature=0.1 for test_tool_choice_required_streaming in the shared tool-choice e2e base.

Test Plan

• python3 -m py_compile e2e_test/chat_completions/test_function_calling.py
• GPU e2e not run locally; this should be validated by the PR e2e matrix, with the Mistral required streaming job rerun a few times to confirm the flake rate drops.

Checklist

• cargo +nightly fmt passes
• cargo clippy --all-targets --all-features -- -D warnings passes
• (Optional) Documentation updated
• (Optional) Please join us on Slack #sig-smg to discuss, review, and merge PRs

Summary by CodeRabbit

Release Notes

This release contains internal test enhancements with no user-visible changes.

• Tests
  • Improved streaming function calling test coverage with adjusted test parameters.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Organization UI

Review profile: ASSERTIVE

Plan: Pro

Run ID: 94e7874f-06e6-4f98-bc0b-31bbbd81bbe9

📥 Commits

Reviewing files that changed from the base of the PR and between 401e666 and a6c1265.

📒 Files selected for processing (1)

• e2e_test/chat_completions/test_function_calling.py

---

📝 Walkthrough

Walkthrough

The streaming chat completions test now explicitly sets temperature=0.1 in its request configuration.

Changes

Streaming Test Configuration

Layer / File(s)	Summary
Streaming test temperature parameter e2e_test/chat_completions/test_function_calling.py	The test_tool_choice_required_streaming method adds temperature=0.1 to the streaming request configuration.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~2 minutes

Poem

A tiny parameter, warm and bright,
temperature=0.1 shines with light,
In the streaming test it now does dwell,
Consistency served with care so well! 🐰✨

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately describes the main change: stabilizing a flaky Mistral streaming test by pinning temperature, which is the primary modification in the changeset.
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 80.00%.
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.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch codex/fix-mistral-required-streaming-flake

---

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

[claude]

Looks good — single-line fix to stabilize a flaky e2e test by pinning temperature, consistent with the non-streaming sibling (temperature=0.2 at line 1037) and the stricter streaming test (temperature=0.1). No issues found.

[gemini-code-assist]

Code Review

This pull request updates the test_tool_choice_required_streaming test case in e2e_test/chat_completions/test_function_calling.py by adding a temperature parameter set to 0.1 to the chat completions request. This change likely improves the determinism of the test output. I have no feedback to provide as there were no review comments.