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/providers/openai.go

@@ -68,6 +68,53 @@ func (p *OpenAIProvider) ChatCompletion(ctx context.Context, req *models.Unified
 	return ParseOpenAIResponse(respJSON, req.Model)
 }
 
+func (p *OpenAIProvider) ImageGeneration(ctx context.Context, req *models.ImageGenerationRequest) (*models.ImageGenerationResponse, error) {
+	body := map[string]interface{}{
+		"prompt": req.Prompt,
+		"model":  req.Model,
+	}
+
+	if req.N != nil {
+		body["n"] = *req.N
+	}
+	if req.Quality != nil {
+		body["quality"] = *req.Quality
+	}
+	if req.ResponseFormat != nil {
+		body["response_format"] = *req.ResponseFormat
+	}
+	if req.Size != nil {
+		body["size"] = *req.Size
+	}
+	if req.Style != nil {
+		body["style"] = *req.Style
+	}
+	if req.User != nil {
+		body["user"] = *req.User
+	}
+
+	resp, err := p.client.R().
+		SetContext(ctx).
+		SetHeader("Authorization", "Bearer "+p.apiKey).
+		SetBody(body).
+		Post(fmt.Sprintf("%s/images/generations", p.config.BaseURL))
+
+	if err != nil {
+		return nil, fmt.Errorf("request failed: %w", err)
+	}
+
+	if !resp.IsSuccess() {
+		return nil, fmt.Errorf("OpenAI image API error (%d): %s", resp.StatusCode(), resp.String())
+	}
+
+	var result models.ImageGenerationResponse
+	if err := json.Unmarshal(resp.Body(), &result); err != nil {
+		return nil, fmt.Errorf("failed to parse response: %w", err)
+	}
+
+	return &result, nil
+}
+
 func (p *OpenAIProvider) ChatCompletionStream(ctx context.Context, req *models.UnifiedRequest) (<-chan *models.ChatCompletionStreamResponse, error) {
 	messagesJSON, err := MessagesToOpenAIJSON(req.Messages)
 	if err != nil {