feat(config): update codex configuration model setting

[HagaSpa]

Background / 背景

Updated the Codex configuration to use the gpt-5-codex model and removed the deprecated model_reasoning_effort setting.

Codex設定を更新し、gpt-5-codexモデルを使用するように変更し、非推奨のmodel_reasoning_effort設定を削除しました。

Changes / 変更内容

• Removed model_reasoning_effort = "high" from .codex/config.toml

• Added model = "gpt-5-codex" to .codex/config.toml

• .codex/config.tomlからmodel_reasoning_effort = "high"を削除

• .codex/config.tomlにmodel = "gpt-5-codex"を追加

Impact scope / 影響範囲

Low impact - configuration change only affects Codex model selection behavior.

影響度低 - 設定変更はCodexのモデル選択動作のみに影響します。

Testing / 動作確認

• Configuration file syntax is valid / 設定ファイルの構文が有効
• Git commit follows conventional commit format / Gitコミットが規約に従っている

🤖 Generated with Claude Code

[claude]

Claude finished @HagaSpa's task —— View job

---

PR Review Complete ✅

Todo List:

• Read and analyze the changed configuration file
• Review code quality and best practices
• Check for potential bugs or issues
• Assess performance considerations
• Evaluate security concerns
• Review test coverage implications
• Provide comprehensive feedback

---

Review Feedback

✅ Code Quality & Best Practices

• Configuration structure: Clean TOML format with proper section organization
• Commit message: Follows conventional commits format (feat(config):)
• Documentation: PR description clearly explains the changes in both English and Japanese
• Change scope: Appropriately focused - single responsibility change

✅ Potential Bugs & Issues

• TOML syntax: Valid syntax confirmed
• Configuration keys: The model specification "gpt-5-codex" appears to be a valid model identifier
• Setting removal: Removing deprecated model_reasoning_effort is appropriate and won't break functionality

✅ Performance Considerations

• Minimal impact: Configuration changes have no runtime performance implications
• Model selection: The gpt-5-codex model should provide good performance for code-related tasks
• No performance regressions: Removal of deprecated settings won't impact performance

✅ Security Concerns

• No sensitive data: Configuration contains only model preferences and feature flags
• Safe values: All configuration values are standard options with no security risks
• No credentials: No API keys or sensitive information exposed

✅ Test Coverage

• Configuration validation: Author confirmed syntax validation ✓
• Low-risk change: Simple configuration update with minimal testing requirements
• Recommendation: Consider adding a basic config validation test if this becomes a frequently modified file

Summary

This is a well-executed, low-risk configuration update. The changes are:

• ✅ Technically sound
• ✅ Properly documented
• ✅ Following project conventions
• ✅ Safe to merge

No issues found - this PR is ready for approval.