feat(llms): add Auxen as an OpenAI-compatible provider

[ai-auxen]

Adds Auxen as a first-class Continue provider. Auxen hosts per-customer dedicated LLM endpoints (Llama 3.1/3.2, Qwen 2.5, Mistral, Gemma 2, Mixtral, Phi-3, Command R) on stable HTTPS URLs with an OpenAI-compatible `/v1/chat/completions` API. Each Auxen instance is a dedicated GPU billed per-minute of runtime, not per-token.

Why a dedicated provider

Auxen is OpenAI-wire-compatible, so users could in principle use the existing `openai` provider with a custom `apiBase`. Registering Auxen as its own provider improves UX in three ways:

1. Appears in the provider dropdown in the config-schema autocomplete
2. Renders provider-specific markdown help (apiBase format, where to get the API key)
3. Lands in the apiKey-required providers list so the JSON schema enforces the right shape

About the per-instance `apiBase`

Unlike most OpenAI-compatible providers that have a single global base URL, every Auxen instance is issued its own URL of the form `https://api.auxen.ai/v1/inst_xxx/v1\`. Users set both `apiBase` and `apiKey` in their config:

```yaml
models:

• name: Auxen Llama 3.1 8B
  provider: auxen
  model: llama-3.1-8b
  apiBase: https://api.auxen.ai/v1/inst_xxx/v1
  apiKey: auxk_...
  ```

Files touched

• `core/llm/llms/Auxen.ts` (new) — thin OpenAI subclass
• `core/llm/llms/index.ts` — register `Auxen` in `LLMClasses`
• `packages/openai-adapters/src/types.ts` — add `"auxen"` to `OpenAIConfigSchema` provider union
• `packages/openai-adapters/src/index.ts` — add `case "auxen"` switch arm
• `extensions/vscode/config_schema.json` — add to main provider enum + matching markdown description, plus the apiKey-required providers list
• `docs/customize/model-providers/more/auxen.mdx` (new) — provider docs page

Companion PRs (Auxen distribution series)

• BerriAI/litellm#28532 — LiteLLM provider
• vercel/ai#15536 — Vercel AI SDK community provider
• langchain-ai/docs#4146 — LangChain docs (Python + JS)
• run-llama/llama_index#21764 — LlamaIndex Python
• npm: `@auxen/ai-sdk-provider` (live)

---

Summary by cubic

Add auxen as an OpenAI-compatible provider with per-instance endpoints, so users can select Auxen in config and use dedicated GPU-backed models. This improves autocomplete, provider help, validation, and adds docs.

• New Features
  • Added Auxen provider (thin OpenAI subclass). Users must set apiBase (per-instance) and apiKey.
  • Registered in LLMClasses and packages/openai-adapters (provider: "auxen").
  • Updated VS Code config_schema.json: provider enum, markdown help, and inclusion in the apiKey-required list.
  • Added docs page: docs/customize/model-providers/more/auxen.mdx.

^{Written for commit 7e00e44. Summary will update on new commits. Review in cubic}

[github-actions]

Thank you for your submission, we really appreciate it. Like many open-source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution. You can sign the CLA by just posting a Pull Request Comment same as the below format.

---

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

---

_{You can retrigger this bot by commenting recheck in this Pull Request. Posted by the CLA Assistant Lite bot.}

[cubic-dev-ai]

No issues found across 6 files

_{Re-trigger cubic}