TINYDOC-3550 - Clarify on-premises AI quick-start provider configuration

[kemister85]

Ticket: TINYDOC-3550

Site: Staging branch

Changes:

• Added an IMPORTANT admonition after the "Launch the AI service" docker run block clarifying that the quick-start command configures a single OpenAI provider, and that adding PROVIDERS, MODELS, or other provider settings to the .env file alone has no effect (only variables passed in docker run reach the service).
• Documented that any other provider — openai-compatible (Databricks, Ollama, vLLM), Azure, Amazon Bedrock, or Google Vertex — requires editing the docker run command to replace PROVIDERS and add a MODELS array, and that MODELS is required for every provider except direct OpenAI, Anthropic, and Google. Cross-referenced the LLM providers page.
• Clarified at the launch step that the docker run command maps the friendly .env names onto the service's own environment variables (for example AI_LICENCE_KEY → LICENSE_KEY, MANAGEMENT_SECRET → ENVIRONMENTS_MANAGEMENT_SECRET_KEY).

---

Pre-checks:

• Branch is correctly prefixed: hotfix/8/TINYDOC-3550
• modules/ROOT/nav.adoc has been updated (if applicable). — not applicable (no new pages)
• Files have been included where required (if applicable). — not applicable
• Files removed have been deleted, not just excluded from the build (if applicable). — not applicable
• Files added for New product features include a release note entry. — not applicable (clarification to existing page)
• Major or minor version changes have updated the supported-versions.adoc table. — not applicable
• Build passes without console errors, warnings, or issues.

---

Review:

• Documentation Team Lead has reviewed.