feat: MCP tool client infrastructure for agent extensibility

Add the full MCP tool pipeline enabling agents to invoke external tools
(like Brave Search) via MCP servers:

- Add ToolRequest/ToolResponse types and mcp-tool topics to @trustgraph/base
- Create McpToolService (FlowProcessor) that connects to external MCP servers
  via @modelcontextprotocol/sdk StreamableHTTP transport
- Add createMcpTool() to wire MCP tools into the agent's ReAct loop
- Implement config-driven tool registration in AgentService with backward-
  compatible fallback to hardcoded tools
- Add tool filtering by group and state (port of Python tool_filter.py)
- Register mcp-tool in gateway dispatcher and export from @trustgraph/flow
- Fix flow restart race condition: skip restart when flow definitions unchanged
- Update seed config with MCP server config and tool definitions
- Add run scripts for MCP tool service and Brave Search MCP server

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

ts/packages/base/src/schema/messages.ts

@@ -304,6 +304,19 @@ export interface CollectionManagementResponse {
   collections?: { user: string; collection: string; name: string; description: string; tags: string[] }[];
 }
 
+// ---------- Tool invocation (MCP tools) ----------
+
+export interface ToolRequest {
+  name: string;
+  parameters: string; // JSON-encoded
+}
+
+export interface ToolResponse {
+  error?: TgError;
+  text?: string;    // Plain text response
+  object?: string;  // JSON-encoded structured response
+}
+
 // ---------- Flow management ----------
 
 // Flow request/response use kebab-case wire format to match the client.

ts/packages/base/src/schema/topics.ts

@@ -60,6 +60,10 @@ export const topics = {
   promptRequest: topic("prompt-request"),
   promptResponse: topic("prompt-response"),
 
+  // MCP tool invocation
+  mcpToolRequest: topic("mcp-tool-request"),
+  mcpToolResponse: topic("mcp-tool-response"),
+
   // Librarian (document management)
   librarianRequest: topic("librarian-request"),
   librarianResponse: topic("librarian-response"),