feat(gateway): 补齐桌面端高级媒体直连接口

新增图片矢量化、视频超分、每日用量、计价与任务隔离能力，并通过环境变量解析平台凭据。

已通过 Go 全量门禁、迁移检查、镜像构建以及 Vectorizer 五格式和 Topaz 3 秒视频真实 DEV 验收。

apps/api/internal/httpapi/openapi_models.go

@@ -210,17 +210,31 @@ type PricingEstimateResponse struct {
 	RequestFingerprint string                   `json:"requestFingerprint" example:"76ef6a537de8e71bd1ca93acadc078dbdbfa9f17e45224e4f9df59f535d2886f"`
 }
 
+type DailyTokenUsageResponse struct {
+	Items     []store.DailyTokenUsage `json:"items"`
+	TokenDays []store.DailyTokenUsage `json:"tokenDays"`
+	Summary   dailyTokenUsageSummary  `json:"summary"`
+	Range     map[string]string       `json:"range"`
+}
+
 type TaskRequest struct {
-	Model      string        `json:"model" example:"gpt-4o-mini"`
-	Messages   []ChatMessage `json:"messages,omitempty"`
-	Input      interface{}   `json:"input,omitempty"`
-	Prompt     string        `json:"prompt,omitempty" example:"A watercolor robot reading a book"`
-	Text       string        `json:"text,omitempty" example:"Hello from EasyAI audio synthesis."`
-	TextFileID string        `json:"text_file_id,omitempty" example:""`
-	VoiceID    string        `json:"voice_id,omitempty" example:"female-shaonv"`
-	Stream     *bool         `json:"stream,omitempty" example:"false"`
-	RunMode    string        `json:"runMode,omitempty" example:"simulation"`
-	MaxTokens  *int          `json:"max_tokens,omitempty" example:"512"`
+	Model          string        `json:"model" example:"gpt-4o-mini"`
+	Messages       []ChatMessage `json:"messages,omitempty"`
+	Input          interface{}   `json:"input,omitempty"`
+	Query          string        `json:"query,omitempty" example:"Which document mentions EasyAI Gateway?"`
+	Documents      []string      `json:"documents,omitempty"`
+	Prompt         string        `json:"prompt,omitempty" example:"A watercolor robot reading a book"`
+	Text           string        `json:"text,omitempty" example:"Hello from EasyAI audio synthesis."`
+	TextFileID     string        `json:"text_file_id,omitempty" example:""`
+	VoiceID        string        `json:"voice_id,omitempty" example:"female-shaonv"`
+	AudioURL       string        `json:"audio_url,omitempty" example:"https://example.com/source-voice.mp3"`
+	DisplayName    string        `json:"display_name,omitempty" example:"EasyAI DEV acceptance"`
+	PromptAudioURL string        `json:"prompt_audio_url,omitempty" example:"https://example.com/prompt-voice.mp3"`
+	PromptText     string        `json:"prompt_text,omitempty" example:"EasyAI voice clone prompt."`
+	PreviewModel   string        `json:"preview_model,omitempty" example:"speech-2.8-hd"`
+	Stream         *bool         `json:"stream,omitempty" example:"false"`
+	RunMode        string        `json:"runMode,omitempty" example:"simulation"`
+	MaxTokens      *int          `json:"max_tokens,omitempty" example:"512"`
 	// MaxCompletionTokens includes visible output and reasoning tokens.
 	MaxCompletionTokens *int `json:"max_completion_tokens,omitempty" example:"512"`
 	MaxOutputTokens     *int `json:"max_output_tokens,omitempty" example:"512"`
@@ -368,6 +382,34 @@ type ImageEditRequest struct {
 	RunMode        string   `json:"runMode,omitempty" example:"simulation"`
 }
 
+type ImageVectorizeSource struct {
+	URL              string `json:"url,omitempty" example:"https://example.com/source.png"`
+	VectorizerTaskID string `json:"vectorizerTaskId,omitempty" example:"9f4d8f3d-5f5f-4bb7-a4be-344a9f930e25"`
+}
+
+type ImageVectorizeRequest struct {
+	Model        string               `json:"model,omitempty" example:"easy-image-vectorizer-1"`
+	Source       ImageVectorizeSource `json:"source"`
+	Format       string               `json:"format,omitempty" example:"svg" enums:"svg,eps,pdf,dxf,png"`
+	MaxColors    int                  `json:"maxColors,omitempty" example:"16" enums:"0,2,4,8,16,32"`
+	CleanupLevel string               `json:"cleanupLevel,omitempty" example:"standard" enums:"low,standard,strong"`
+}
+
+type VideoUpscaleRequest struct {
+	Model            string  `json:"model,omitempty" example:"easy-proteus-standard-4"`
+	VideoURL         string  `json:"video_url" example:"https://example.com/source.mp4"`
+	Operation        string  `json:"operation,omitempty" example:"upscale" enums:"upscale"`
+	TargetResolution string  `json:"target_resolution,omitempty" example:"1080p"`
+	OutputWidth      int     `json:"output_width,omitempty" example:"1920"`
+	OutputHeight     int     `json:"output_height,omitempty" example:"1080"`
+	PreserveAudio    *bool   `json:"preserve_audio,omitempty" example:"true"`
+	Duration         float64 `json:"duration,omitempty" example:"3"`
+	SourceResolution string  `json:"source_resolution,omitempty" example:"480p"`
+	SourceFrameRate  float64 `json:"source_frame_rate,omitempty" example:"24"`
+	TargetFrameRate  float64 `json:"target_frame_rate,omitempty" example:"24"`
+	SlowMotionRate   float64 `json:"slow_motion_rate,omitempty" example:"1"`
+}
+
 type VideoGenerationRequest struct {
 	Model      string `json:"model" example:"video-model"`
 	Prompt     string `json:"prompt" example:"A cinematic drone shot over mountains"`