language_models: Add images support for Ollama vision models (#31883)

Ollama supports vision to process input images. This PR adds support for
same. I have tested this with gemma3:4b and have attached the screenshot
of it working.

<img width="435" alt="image"
src="https://github.com/user-attachments/assets/5f17d742-0a37-4e6c-b4d8-05b750a0a158"
/>


Release Notes:

- Add image support for [Ollama vision models](https://ollama.com/search?c=vision)

crates/agent_settings/src/agent_settings.rs

@@ -372,6 +372,7 @@ impl AgentSettingsContent {
                                 None,
                                 None,
                                 Some(language_model.supports_tools()),
+                                Some(language_model.supports_images()),
                                 None,
                             )),
                             api_url,