feat: add litellm as AI provider gateway

[prodmanpd]

Description

Adds LiteLLM as a first-class model provider (provider: litellm). LiteLLM is an open-source, self-hosted AI gateway that fronts 100+ LLM providers behind one OpenAI-compatible API. LiteLLM extends the OpenAI class (mirroring the existing OpenRouter / ClawRouter gateway providers) and defaults apiBase to the local proxy (http://localhost:4000/v1/). A user points apiBase at their proxy with a key and either names a model or uses model: AUTODETECT to discover everything the proxy serves via GET /v1/models.

Changes:

• core/llm/llms/LiteLLM.ts - new provider (providerName = "litellm", default proxy apiBase, reasoning-field support); registered in core/llm/llms/index.ts.
• core/llm/autodetect.ts - added to PROVIDER_HANDLES_TEMPLATING and PROVIDER_SUPPORTS_IMAGES (a proxy can route to vision models).
• core/llm/toolSupport.ts - tool-support detection by routed model name.
• gui/src/pages/AddNewModel/configs/providers.ts - "Add Model" onboarding entry (API-key + Base URL inputs, then AUTODETECT).
• Docs: more/litellm.mdx (incl. a model: AUTODETECT example) + overview table row and nav entry.

AI Code Review

• Team members only: AI review runs automatically when PR is opened or marked ready for review
• Team members can also trigger a review by commenting @continue-review

Checklist

• I've read the contributing guide
• The relevant docs, if any, have been updated or created
• The relevant tests, if any, have been updated or created

Tests

• Provider unit tests: core/llm/llms/LiteLLM.vitest.ts (5 passed) and a litellm block added to core/llm/toolSupport.test.ts.
• Whole gui package: npm run tsc:check -> 0 errors (including the new Add-Model entry); npm run test (vitest) -> 37 files / 372 tests passed. npx tsc --noEmit on core/ is clean.
• Live end-to-end through a running LiteLLM proxy - drove the real LiteLLM.chat(...) (-> streamChat -> OpenAI adapter -> proxy):

  provider=litellm  model=gpt-4.1-mini  RESPONSE: LITELLM_OK   (proxy -> Azure OpenAI)
  

• Model discovery: LiteLLM extends OpenAI, and OpenAI.listModels() fetches GET /v1/models, so model: AUTODETECT auto-populates the models the proxy serves.

[github-actions]

All contributors have signed the CLA ✍️ ✅
_{Posted by the CLA Assistant Lite bot.}

[prodmanpd]

I have read the CLA Document and I hereby sign the CLA