[AAASM-5568] 🚨 (python-sdk): Clear 44 SonarCloud code smells

[Chisanan232]

Description

Clears the 44 SonarCloud code smells on python-sdk main down toward the quality target (<5). These surfaced only after the SonarCloud main branch was corrected from the stale master snapshot to main and re-scanned (AAASM-5562) — the old snapshot falsely reported 3. Nearly all are in test/; one is production code.

Fixes by rule:

• S8997 ×31 — use pytest's monkeypatch fixture for temporary attribute/env/context modifications instead of manual set-and-restore (auto-restores; manual teardown removed). Across test_runner_spawn_patch, test_crewai_spawn_context, test_edge_emission, test_pydantic_ai_spawn_patch, test_registry, bench latency contracts, etc.
• S9083 ×8 — remove empty parentheses from no-arg decorators (@pytest.fixture() → @pytest.fixture) in cli/conftest and native_core.
• S9073 ×3 — split composite assert A and B into separate assertions.
• S5778 ×1 — hoist argument construction out of the pytest.raises block so only the single throwing call remains inside.
• S3776 ×1 — reduce cognitive complexity 18→<15 in agent_assembly/adapters/openai_agents/patch.py by extracting the tool pre-execution check (only production-code change; pure refactor).

Type of Change

• ♻️ Refactoring

Breaking Changes

• No

Related Issues

• Related JIRA ticket: AAASM-5568 (Epic AAASM-5567)

Testing

• Unit tests added/updated — the monkeypatch conversions restructure test internals; all covered
• Integration tests added/updated
• Manual testing performed
• No tests required

Verification (worktree, uv): ruff check . clean; full uv run pytest = 1204 passed, 16 skipped (skips are env-gated: native _core not built, optional adapter libs absent), 0 failures. mypy shows only the 4 pre-existing _core/grpc-stub import errors that are identical on main (not introduced here). SonarCloud's cognitive-complexity metric is scanner-only, so the S3776 reduction is confirmed by this PR's SonarCloud analysis.

Checklist

• Code follows project style guidelines
• Self-review completed
• Comments added for complex logic
• Documentation updated if needed — N/A
• All tests passing

🤖 Generated with Claude Code

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
100.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[Chisanan232]

✅ Review — LGTM (Claude Code)

Reviewed against the full standard: requirements/logic, implementation, bugs/regressions, goal match, CI.

Goal match — Clears the 44 code smells surfaced by the corrected main re-scan (AAASM-5568), targeting <5. Scope is exactly the flagged rules; no unrelated changes.

Implementation — Correct and idiomatic:

• S8997 ×31: manual attr/env/context set-and-restore replaced with pytest monkeypatch (auto-restore; manual teardown removed). This is the sanctioned pattern.
• S9083 ×8: empty parens dropped only from genuinely no-arg decorators.
• S9073 ×3: top-level assert A and B split; messages preserved.
• S5778: arg construction hoisted out of pytest.raises so only the throwing call remains inside.
• S3776: patch.py tool pre-execution check extracted — production refactor, no behavior change.

No regression — Verified locally (Node/uv, worktree): ruff check . clean; full uv run pytest = 1204 passed, 16 skipped (skips env-gated: native _core unbuilt, optional adapter libs absent), 0 failures. mypy shows only the 4 pre-existing _core/grpc-stub import errors identical on main — not introduced here.

CI — 19 pass / 7 skip / 0 fail. SonarCloud PR analysis: open issues = 0 → all 44 resolved, no new issue introduced.

No missing requirements or logic gaps. 🟢