feat: add llmman as a local model provider

[ericcurtin]

What this PR does / why we need it?

Adds llmman, a local model runner that serves the Ollama API on port 17434, as a model provider.

Summary of your change

• apps/models_provider/impl/llmman_model_provider/: LlmmanModelProvider subclasses OllamaModelProvider and reuses the Ollama credentials, OllamaChatModel, OllamaEmbedding, /api/tags listing and /api/pull download as-is. Reranker and vision are left for a follow-up.
• ModelProvideConstants: registers model_llmman_provider next to Ollama.
• ui: adds it to the local-provider grouping in Provider.vue and the static list in provider-data.ts, with a neutral terminal glyph as icon (force-added like the other gitignored impl/*/icon/ files).
• Default model names are suggestions only; users enter http://localhost:17434 as the API URL.

Testing: ruff check and py_compile on the new package pass; provider imports in a Django-configured interpreter and resolves LLM/EMBEDDING models to the Ollama classes; vue-tsc --build and eslint clean on the two UI files. Not run against a live llmman server.

Please indicate you've done the following:

• Made sure tests are passing and test coverage is added if needed. (No test runner is configured in this repo.)
• Made sure commit message follow the rule of Conventional Commits specification.
• Considered the docs impact and opened a new docs issue or PR with docs changes if needed. (Docs live outside this repo.)

AI-assisted, reviewed before submitting.