[Feature]: docs: openai-compatible hosted /v1 example

Which destkop app does this feature request relate to?

• Agent-TARS (cli, server, agent, tool etc.)
• UI-TARS Desktop

What problem does this feature solve?

The CLI and model provider pages already cover custom OpenAI-compatible hosts. Three details are easy to get wrong:

1. baseURL / --model.baseURL must include /v1. Origin-only URLs miss the chat path this host expects.
2. id / --model.id is the catalog id. Do not prefix it with openai/ (that prefix is for OpenRouter / LiteLLM).
3. Prefer provider: "openai-compatible" (or an unknown name, which already defaults to that). Do not treat this as a UI-TARS Desktop branded-provider request.

PZERO is a live OpenAI-compatible host. Catalog: GET https://api.pzero.studio/v1/models. Default text id: deepseek-v4-flash. Prepaid Bearer key (pzero_…). Chat Completions is the working path.

What does the proposed features look like?

No first-party provider asked. A worked CLI example for searchers is enough:

npx agent-tars \
  --model.provider openai-compatible \
  --model.id deepseek-v4-flash \
  --model.apiKey pzero_… \
  --model.baseURL https://api.pzero.studio/v1

Config equivalent:

{
  provider: "openai-compatible",
  id: "deepseek-v4-flash",
  apiKey: "pzero_…",
  baseURL: "https://api.pzero.studio/v1",
}

Do not append /chat/completions to baseURL. Do not prefix id with openai/. If a named vendor subsection is too much, the gotchas above are the docs gap.