Add configurable evolution output token limit

[cursor]

Summary

Adds a configurable max output token limit for evolution model calls, defaulting to 32,768 instead of the previous hardcoded 65,000. The preference is stored with the existing evolution limits, exposed in Settings -> AI Models -> Evolution Limits, used for OpenAI-compatible/vLLM and Ollama evolution calls, and available from the CLI via --max-output-tokens.

Also adds clearer context-window guidance when providers reject a request because the configured output token limit is too high.

Fixes ENG-519

Test Plan

• bun install
• bun -F native build
• bun -F native lint
• bun -F native test:storybook -- ai-models-tab
• cargo +stable test providers::tests::context_window_errors_suggest_token_setting --manifest-path Cargo.toml (blocked in this Linux environment by missing gdk-3.0 Tauri system library)

Docs

• Docs updated (README self-hosted vLLM max output token configuration)
• No docs update needed

  

---

^{Need help on this PR? Tag @codesmith with what you need. Autofix is disabled.}

[github-actions]

	Warnings
⚠️	PR is marked WIP / draft — do not merge until ready for review.
⚠️	Please assign this PR to someone (usually yourself).

	Messages
📖	Docs companion PR linked — thank you.

📋 PR Overview

Lines changed	255 (+226 / -29)
Files	0 added, 17 modified, 0 deleted
Draft / WIP	yes
Has Test Plan	yes
New UI components	no
New Storybook stories	no
New Rust modules	no
New TS source files	no
New tests	no
package.json touched	no
Cargo.toml touched	no
Infra / CI touched	no

🔬 Coverage

Report	Lines	Statements	Functions	Branches
apps/native/coverage/coverage-summary.json	17.9%	17.9%	31.0%	54.1%

Generated by 🚫 dangerJS against 4578518