gpt-4o-mini does handle images

Permit sending images to gpt-4o-mini.

Also, change `DescribeImagesUsingChatGPT` example to use the default model (now gpt-4o-mini), and while we are editing it anyway, remove the `addpath` and `loadenv` lines and use the newer `wrapText` utility function.

Author: ccreutzi

+llms/+openai/validateMessageSupported.m

@@ -5,7 +5,9 @@
 
     % only certain models support image generation
     if iscell(message.content) && any(cellfun(@(x) isfield(x,"image_url"), message.content))
-        if ~ismember(model,["gpt-4-turbo","gpt-4-turbo-2024-04-09","gpt-4o","gpt-4o-2024-05-13"]) 
+        if ~ismember(model,["gpt-4-turbo","gpt-4-turbo-2024-04-09",...
+            "gpt-4o-mini","gpt-4o-mini-2024-07-18",...
+            "gpt-4o","gpt-4o-2024-05-13"]) 
          error("llms:invalidContentTypeForModel", ...
                llms.utils.errorMessageCatalog.getMessage("llms:invalidContentTypeForModel", "Image content", model));
         end

examples/DescribeImagesUsingChatGPT.mlx

@@ -190,7 +190,7 @@ function assignValueToProperty(property, value)
         end
 
         function invalidGenerateInputforModel(testCase)
-            chat = openAIChat(APIKey="this-is-not-a-real-key");
+            chat = openAIChat(APIKey="this-is-not-a-real-key",Model="gpt-3.5-turbo");
             image_path = "peppers.png";
             emptyMessages = messageHistory;
             inValidMessages = addUserMessageWithImages(emptyMessages,"What is in the image?",image_path);