feat(ai-core): real vision image analysis

- llm: Adapter.Vision sends multimodal chat requests (text + image_url,
  data URLs supported) and logs the call as purpose 'vision'
- gateway /images/analyze now accepts image_url or image_base64 and
  returns real analysis instead of a placeholder message
- main wires the default LLM adapter as the vision analyzer
- httptest coverage for the vision request/response shape

backend/services/ai-core/llm/llm.go

@@ -76,6 +76,42 @@ type CallRecord struct {
 	Error          string
 }
 
+// Vision sends a prompt plus an image (URL or data URL) to a multimodal model
+// and returns the analysis text.
+func (a *Adapter) Vision(ctx context.Context, prompt, imageRef string) (*model.LLMResponse, error) {
+	start := time.Now()
+	body := map[string]any{
+		"model": a.cfg.Model,
+		"messages": []map[string]any{
+			{
+				"role": "user",
+				"content": []map[string]any{
+					{"type": "text", "text": prompt},
+					{"type": "image_url", "image_url": map[string]string{"url": imageRef}},
+				},
+			},
+		},
+	}
+
+	resp := &model.LLMResponse{Model: a.cfg.Model}
+	err := a.doRequest(ctx, "POST", "/chat/completions", body, func(data map[string]any) error {
+		if choices, ok := data["choices"].([]any); ok && len(choices) > 0 {
+			choice := choices[0].(map[string]any)
+			if msg, ok := choice["message"].(map[string]any); ok {
+				if content, ok := msg["content"].(string); ok { resp.Content = content }
+			}
+			if reason, ok := choice["finish_reason"].(string); ok { resp.FinishReason = reason }
+		}
+		if usage, ok := data["usage"].(map[string]any); ok {
+			if total, ok := usage["total_tokens"].(float64); ok { resp.TokensUsed = int(total) }
+		}
+		return nil
+	})
+
+	a.recordCall(start, "vision", resp.TokensUsed, 0, err)
+	return resp, err
+}
+
 // ModelName returns the configured model name.
 func (a *Adapter) ModelName() string { return a.cfg.Model }
 

backend/services/ai-core/llm/llm_test.go

@@ -0,0 +1,37 @@
+package llm
+
+import (
+	"context"
+	"io"
+	"net/http"
+	"net/http/httptest"
+	"strings"
+	"testing"
+)
+
+func TestVision(t *testing.T) {
+	srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
+		body, _ := io.ReadAll(r.Body)
+		s := string(body)
+		if !strings.Contains(s, `"type":"image_url"`) || !strings.Contains(s, "data:image/jpeg;base64,AAAA") {
+			t.Errorf("unexpected vision request body: %s", s)
+		}
+		w.Write([]byte(`{"choices":[{"message":{"content":"图片里是一只猫"},"finish_reason":"stop"}],"usage":{"total_tokens":42}}`))
+	}))
+	defer srv.Close()
+
+	cfg := DefaultConfig()
+	cfg.BaseURL = srv.URL
+	a := NewAdapter(cfg)
+
+	resp, err := a.Vision(context.Background(), "这是什么？", "data:image/jpeg;base64,AAAA")
+	if err != nil {
+		t.Fatalf("Vision: %v", err)
+	}
+	if resp.Content != "图片里是一只猫" {
+		t.Errorf("content = %q", resp.Content)
+	}
+	if resp.TokensUsed != 42 {
+		t.Errorf("tokens = %d", resp.TokensUsed)
+	}
+}