feat(settings): add Requesty provider preset

[Thibaultjaigu]

Summary

Adds Requesty (https://requesty.ai) as a provider preset in AI settings, next to OpenRouter. Requesty is an OpenAI compatible LLM gateway, so it goes through the existing generic openai path with no backend changes, keeping the provider agnostic design. This matches what was merged for DocKit in geek-fun/dockit#462.

Changes

• src/views/setting/provider-form-dialog.vue: requesty preset (apiCompatibility: 'openai', base URL https://router.requesty.ai/v1), placed right after OpenRouter
• CHANGELOG.md: one line under Unreleased / Added

How to test

1. Settings > AI > Add provider, pick Requesty, paste a key from https://app.requesty.ai/api-keys
2. Sync models or test connection (uses GET /v1/models), then chat with a model such as openai/gpt-4o-mini

Validation

• npm run lint:check: clean
• vue-tsc --noEmit: clean
• npm run test:ci: 28 suites, 533 tests passed
• Live check against Requesty with the same requests the harness sends for an openai compatible preset: GET /v1/models 200 (includes openai/gpt-4o-mini), streamed POST /v1/chat/completions with openai/gpt-4o-mini 200 with the expected reply. I did not build the full Tauri app locally.

Disclosure: I work at Requesty. Happy to adjust anything to match project conventions.