fix: personality LLM now respects the selected refinement model

[aajinn]

Bug: Personality LLM (Speak in character / Compose) always loaded Qwen3 0.6B regardless of the user's refinement model selection

• Root cause: rewrite_as_profile and compose_as_profile were called without model_size, so they fell back to backend.model_size (default "0.6B")
• Fix: Both call sites now read saved.llm_model from capture_settings via settings_service.get_capture_settings(db) and pass it as model_size
• Reference the issue [Bug] Refinement model selection ignores user choice, always loads Qwen3 0.6B instead of selected 1.7B/4B

Summary by CodeRabbit

• Bug Fixes
  • Fixed personality-based features to respect your selected LLM model configuration. The "Speak in character" rewrite and Compose action now use your chosen model instead of defaulting to a fixed option, ensuring consistent refinement model usage across all features.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro

Run ID: 9cb110d3-838c-47c1-8a51-fa82b2236868

📥 Commits

Reviewing files that changed from the base of the PR and between b35b909 and 34153e7.

📒 Files selected for processing (3)

• CHANGELOG.md
• backend/routes/generations.py
• backend/routes/profiles.py

---

📝 Walkthrough

Walkthrough

This PR wires personality-driven features to respect the user-configured LLM model setting. The /generate rewrite and /compose endpoints now fetch capture settings from the database and pass the selected model size to personality functions, replacing implicit defaults.

Changes

Personality LLM Model Configuration

Layer / File(s)	Summary
Personality rewrite respects configured LLM model backend/routes/generations.py	Imports settings_service and modifies the personality rewrite path in /generate to fetch capture settings and pass model_size=saved.llm_model to personality.rewrite_as_profile.
Personality compose respects configured LLM model backend/routes/profiles.py	Imports settings_service and modifies the compose path in /profiles/{profile_id}/compose to fetch capture settings and pass model_size=saved.llm_model to personality.compose_as_profile.
Changelog documentation CHANGELOG.md	Documents that personality LLM now respects the selected refinement model in Unreleased → Bug Fixes.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

Poem

A rabbit refines the character's voice,
Now honoring the model of choice,
No hardcoded defaults in sight,
Settings-aware, the rewrite runs right,
Personality speaks with precision tonight! 🐰✨

🚥 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 PR title directly and accurately summarizes the main change: the personality LLM now respects the selected refinement model instead of always using a default.
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 unit tests (beta)

• Create PR with unit tests

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

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