restore model load duration on generate response (#1524)

* restore model load duration on generate response

- set model load duration on generate and chat done response
- calculate createAt time when response created

* remove checkpoints predict opts

* Update routes.go

llm/llama.go

@@ -548,17 +548,12 @@ const maxBufferSize = 512 * format.KiloByte
 const maxRetries = 6
 
 type PredictOpts struct {
-	Prompt           string
-	Format           string
-	Images           []api.ImageData
-	CheckpointStart  time.Time
-	CheckpointLoaded time.Time
+	Prompt string
+	Format string
+	Images []api.ImageData
 }
 
 type PredictResult struct {
-	CreatedAt          time.Time
-	TotalDuration      time.Duration
-	LoadDuration       time.Duration
 	Content            string
 	Done               bool
 	PromptEvalCount    int
@@ -681,16 +676,12 @@ func (llm *llama) Predict(ctx context.Context, predict PredictOpts, fn func(Pred
 
 				if p.Content != "" {
 					fn(PredictResult{
-						CreatedAt: time.Now().UTC(),
-						Content:   p.Content,
+						Content: p.Content,
 					})
 				}
 
 				if p.Stop {
 					fn(PredictResult{
-						CreatedAt:     time.Now().UTC(),
-						TotalDuration: time.Since(predict.CheckpointStart),
-
 						Done:               true,
 						PromptEvalCount:    p.Timings.PromptN,
 						PromptEvalDuration: parseDurationMs(p.Timings.PromptMS),