feat(image): pass reference images to GPT image model for edits

Server: forward canvas/reference images to the image model so extract,
continue, and edit requests stay faithful to the source product.
- Add ImageRequestOptions.Images and route requests with inputs to the
  /v1/images/edits endpoint, via multipart upload (default) or JSON
  image_url payload, selectable through Agent.Image.InputImageTransport.
- Extract reference URLs from prompt directives, inline @image tokens,
  and chat message image/mask contents; fold them into the idempotency
  key and prepend a reference-fidelity instruction to the prompt.

Frontend:
- Shrink CanvasToast to a compact top pill.
- Paste an internal transparent clipboard node instead of re-uploading
  when it matches the clipboard image.
- Keep agent-project polling alive on stream error while streaming.

server/internal/config/config.go

@@ -123,10 +123,11 @@ type AgentConfig struct {
 }
 
 type ImageGenerationConfig struct {
-	BaseURL        string `json:",optional"`
-	APIKey         string `json:",optional"`
-	APIKeyEnv      string `json:",default=GPT_IMAGE_API_KEY"`
-	TimeoutSeconds int    `json:",default=600"`
+	BaseURL             string `json:",optional"`
+	APIKey              string `json:",optional"`
+	APIKeyEnv           string `json:",default=GPT_IMAGE_API_KEY"`
+	InputImageTransport string `json:",default=file"`
+	TimeoutSeconds      int    `json:",default=600"`
 }
 
 type TextExtractionConfig struct {