feat(mcp): add context length guard to prevent oversized requests

* feat: add X-Client-ID header for claw402 monitoring

* feat(mcp): add context length guard to prevent oversized requests

- Add MaxContext field to Config (default 0 = no limit)
- Add WithMaxContext() option for setting model context limits
- Add context_guard.go: token estimation + message truncation
- Integrate guard into both BuildMCPRequestBody and BuildRequestBodyFromRequest
- Support both map[string]string and map[string]any message formats
- Truncates oldest non-system messages when estimated tokens exceed limit
- Always preserves system messages and keeps at least 1 non-system message
- Logs warning when truncation occurs for debugging

Usage: mcp.NewDeepSeekClient(mcp.WithMaxContext(131072))

mcp/client.go

@@ -227,6 +227,16 @@ func (client *Client) BuildMCPRequestBody(systemPrompt, userPrompt string) map[s
 		"content": userPrompt,
 	})
 
+	// Guard: truncate messages if they would exceed the model's context window
+	if client.Cfg.MaxContext > 0 {
+		truncated, removed := truncateMessages(messages, client.Cfg.MaxContext, client.MaxTokens)
+		if removed > 0 {
+			client.Log.Warnf("⚠️  [%s] Context guard: truncated %d oldest messages to fit within %d token limit",
+				client.String(), removed, client.Cfg.MaxContext)
+			messages = truncated
+		}
+	}
+
 	// Build request body
 	requestBody := map[string]interface{}{
 		"model":       client.Model,
@@ -575,6 +585,20 @@ func (client *Client) BuildRequestBodyFromRequest(req *Request) map[string]any {
 		messages = append(messages, m)
 	}
 
+	// Guard: truncate messages if they would exceed the model's context window
+	maxOut := client.MaxTokens
+	if req.MaxTokens != nil {
+		maxOut = *req.MaxTokens
+	}
+	if client.Cfg.MaxContext > 0 {
+		truncated, removed := truncateMessagesAny(messages, client.Cfg.MaxContext, maxOut)
+		if removed > 0 {
+			client.Log.Warnf("⚠️  [%s] Context guard: truncated %d oldest messages to fit within %d token limit",
+				client.String(), removed, client.Cfg.MaxContext)
+			messages = truncated
+		}
+	}
+
 	// Build basic request body
 	requestBody := map[string]interface{}{
 		"model":    req.Model,