Add support for integers and tool configuration in schema handling

This update introduces support for `jsonschema.Integer` types and updates the logic to handle nested items in schemas. Added a new default error log for unknown types using `slog.Error`. Also, integrated tool configuration with a `FunctionCallingConfig` when `dontRequireTool` is false.

response.go

@@ -1,6 +1,8 @@
 package go_llm
 
-import "github.com/sashabaranov/go-openai"
+import (
+	"github.com/openai/openai-go"
+)
 
 type ResponseChoice struct {
 	Index   int
@@ -30,24 +32,34 @@ func (r ResponseChoice) toRaw() map[string]any {
 	return res
 }
 
-func (r ResponseChoice) toChatCompletionMessages() []openai.ChatCompletionMessage {
-	var res = openai.ChatCompletionMessage{
-		Role:    openai.ChatMessageRoleAssistant,
-		Content: r.Content,
-		Refusal: r.Refusal,
+func (r ResponseChoice) toChatCompletionMessages(_ string) []openai.ChatCompletionMessageParamUnion {
+	var as openai.ChatCompletionAssistantMessageParam
+
+	if r.Name != "" {
+		as.Name = openai.String(r.Name)
+	}
+	if r.Refusal != "" {
+		as.Refusal = openai.String(r.Refusal)
+	}
+
+	if r.Content != "" {
+		as.Content.OfString = openai.String(r.Content)
 	}
 
 	for _, call := range r.Calls {
-		res.ToolCalls = append(res.ToolCalls, openai.ToolCall{
-			ID:   call.ID,
-			Type: openai.ToolTypeFunction,
-			Function: openai.FunctionCall{
+		as.ToolCalls = append(as.ToolCalls, openai.ChatCompletionMessageToolCallParam{
+			ID: call.ID,
+			Function: openai.ChatCompletionMessageToolCallFunctionParam{
 				Name:      call.FunctionCall.Name,
 				Arguments: call.FunctionCall.Arguments,
 			},
 		})
 	}
-	return []openai.ChatCompletionMessage{res}
+	return []openai.ChatCompletionMessageParamUnion{
+		{
+			OfAssistant: &as,
+		},
+	}
 }
 
 func (r ResponseChoice) toInput() []Input {