feat: add image generation for OpenAI DALL-E and Gemini Imagen

New `/v1/images/generations` endpoint proxies DALL-E 2/3 (OpenAI)
and Imagen 3 (Gemini). Same auth/logging as chat completions.

- Add ImageGenerationRequest/Response models
- Extend Provider interface with ImageGeneration()
- OpenAI: forward to /v1/images/generations
- Gemini: call /v1beta/models/{model}:predict, map OpenAI params
- Circuit breaker wraps image gen like chat completions
- Model routing: dall-e* -> openai, imagen*/gemini* -> gemini
- Unsupported providers (deepseek/moonshot/grok/ollama) return error
- Fix pre-existing CachedContentTokenCount bug in StreamGemini

internal/models/models.go

@@ -210,6 +210,30 @@ func (i *ImageInput) ToBase64() (string, string, error) {
 	return "", "", fmt.Errorf("empty image input")
 }
 
+// Image Generation (DALL-E, Imagen)
+
+type ImageGenerationRequest struct {
+	Model          string  `json:"model"`
+	Prompt         string  `json:"prompt"`
+	N              *uint32 `json:"n,omitempty"`
+	Quality        *string `json:"quality,omitempty"`
+	ResponseFormat *string `json:"response_format,omitempty"`
+	Size           *string `json:"size,omitempty"`
+	Style          *string `json:"style,omitempty"`
+	User           *string `json:"user,omitempty"`
+}
+
+type ImageGenerationResponse struct {
+	Created int64       `json:"created"`
+	Data    []ImageData `json:"data"`
+}
+
+type ImageData struct {
+	URL           string `json:"url,omitempty"`
+	B64JSON       string `json:"b64_json,omitempty"`
+	RevisedPrompt string `json:"revised_prompt,omitempty"`
+}
+
 // AuthInfo for context
 type AuthInfo struct {
 	Token    string