Request OpenAI compatible endpoint in Advanced Paste

### Description of the new feature / enhancement

Instead of limiting to ollama endpoints, enable the usage of OpenAI compatible endpoints so I can use any self selected inference provider or the local llama.cpp.

### Scenario when this would be used?

- When you have a provider like qwen or zai, which is not in ollama format
- When you use another local inference provider like llama.cpp or lmstudio

### Supporting information

_No response_

Start by locating Advanced Paste's existing Ollama endpoint integration and review the requested OpenAI-compatible endpoint scenarios, including llama.cpp and LM Studio. Determine the supported endpoint configuration and verify that providers such as Qwen or Zai can be used without the Ollama format; the issue names no files or tests.

Written by the indexing model from the issue text.