fix stream usage

llm/openai_compatible/chat.go

@@ -14,6 +14,8 @@ import (
 	"opencatd-open/llm"
 	"os"
 	"strings"
+
+	"github.com/sashabaranov/go-openai"
 )
 
 // https://learn.microsoft.com/en-us/azure/ai-services/openai/api-version-deprecation#latest-preview-api-releases
@@ -130,6 +132,7 @@ func (o *OpenAICompatible) Chat(ctx context.Context, chatReq llm.ChatRequest) (*
 
 func (o *OpenAICompatible) StreamChat(ctx context.Context, chatReq llm.ChatRequest) (chan *llm.StreamChatResponse, error) {
 	chatReq.Stream = true
+	chatReq.StreamOptions = &openai.StreamOptions{IncludeUsage: true}
 	dst, err := utils.StructToMap(chatReq)
 	if err != nil {
 		return nil, err