[YQHD711]

Adds an optional helper on the Full Parameter List that translates English parameter descriptions using any OpenAI-compatible chat-completions endpoint.

• Utilities/LLMTranslationService.cs - endpoint-agnostic client. Defaults to the standard OpenAI chat-completions URL; any compatible base URL can be configured (OpenAI, Azure OpenAI, local Ollama / vLLM / LM Studio, ...). A candidate-URL fallback tries both /v1/chat/completions and /chat/completions so vendor path conventions are not hard coded.
• GCSViews/ConfigurationView/LLMConfigDialog.cs - settings dialog (endpoint, key, model, temperature, custom system prompt, connection test).
• GCSViews/ConfigurationView/ConfigRawParams.cs - toolbar buttons and the right-click entry.

Design notes:

• Inert until configured. Nothing leaves the machine until the user sets an endpoint and key; IsConfigured is false with an empty key and TranslateAsync returns the input unchanged.
• The API key is stored through the existing Settings mechanism and is never logged (debug output records only its length).
• Parameter descriptions are sent to the configured third-party endpoint, which is why the feature is opt-in.
• All user-visible strings go through the existing resx mechanism.

Testing: built successfully; exercised single-row translate, translate-all with progress, stop mid-batch, connection test pass/fail, and the no-key path.

[YQHD711]

Hi, first-time contributor here. The DotNet Build workflow on this PR is still showing as pending rather than running - it looks like it is waiting for approval to run on a fork PR from a first-time contributor. Could a maintainer approve the workflow when convenient? Happy to adjust anything about the change in the meantime.

This one adds an optional, opt-in LLM translation helper for the parameter list. If you would rather not carry a network-dependent feature like this in the main application, please just say so and I will withdraw it - I would rather ask now than waste your review time.