feat: refactor token estimation logic

- Introduced new OpenAI text models in `common/model.go`.
- Added `IsOpenAITextModel` function to check for OpenAI text models.
- Refactored token estimation methods across various channels to use estimated prompt tokens instead of direct prompt token counts.
- Updated related functions and structures to accommodate the new token estimation approach, enhancing overall token management.

service/usage_helpr.go

@@ -23,8 +23,7 @@ func ResponseText2Usage(c *gin.Context, responseText string, modeName string, pr
 	common.SetContextKey(c, constant.ContextKeyLocalCountTokens, true)
 	usage := &dto.Usage{}
 	usage.PromptTokens = promptTokens
-	ctkm := CountTextToken(responseText, modeName)
-	usage.CompletionTokens = ctkm
+	usage.CompletionTokens = EstimateTokenByModel(modeName, responseText)
 	usage.TotalTokens = usage.PromptTokens + usage.CompletionTokens
 	return usage
 }