feat(workflows): add OpenAI-compatible LLM block for custom endpoints

[grzegorz-roboflow]

What does this PR do?

Block that sends chat completion requests to any OpenAI-compatible API (vLLM, Ollama, local Qwen, etc). User provides base URL + model name. Prompt uses email-block-style {{ $parameters.xxx }} template params. Image params (JPEG bytes from Image Stack or WorkflowImageData) auto-base64-encoded and sent as vision content parts.

Type of Change

• New feature (non-breaking change that adds functionality)

Testing

Tests added

Checklist

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have commented my code where necessary, particularly in hard-to-understand areas
• My changes generate no new warnings or errors
• I have updated the documentation accordingly (if applicable)

Additional Context

N/A