fix(cursor): stop AgentService empty turns and silent tool hangs

Cursor-hosted models (cu/composer-2.5, cu/cursor-grok-*, cu/default) returned
HTTP 200 with an empty turn, or hung, whenever a client sent tools.

- Fold system prompts into the current user message. custom_system_prompt
  (RunRequest field 8) makes AgentService return an empty turn.
- Send ModelDetails (field 3); thinking variants (Composer, Grok, *-thinking)
  return an empty turn when only requested_model (field 9) is set.
- Route tool-call history and declared tool schemas through AgentService:
  encode OpenAI tools into mcp_tools (field 4), decode McpArgs and emit real
  tool_calls with finish_reason tool_calls.
- Map Composer  thinking / Grok thinking_delta (field 4) into visible content
  instead of dropping the answer with the unsigned reasoning.
- Ack request_context without echoing MCP tools (double-advertise stalls the
  HTTP/2 stream) and ack kv_server_message so the run proceeds.
- Reject IDE builtin execs instead of failing the turn, so the model can
  continue with MCP tools or a text answer.
- Add google.protobuf.Value / MCP encoders and a FIXED64 branch to
  encodeField in cursorProtobuf.js.

RTK now compresses the source-format body before translation for cursor only:
its translator rewrites role:tool into user XML, so the post-translate pass
missed those tool results. Every other provider keeps the post-translate pass
unchanged.

tests/unit/cursor-agent-proto.test.js

@@ -246,6 +246,15 @@ describe("Cursor AgentService executor helpers (cursor.js)", () => {
       const run = decodeMessage(clientMsg.get(1)[0].value);
       expect(run.has(2)).toBe(true); // action
       expect(run.has(9)).toBe(true); // requested_model
+      // custom_system_prompt (field 8) makes AgentService return an empty turn.
+      expect(run.has(8)).toBe(false);
+      expect(run.has(3)).toBe(true); // ModelDetails — required for thinking variants
+      const action = decodeMessage(run.get(2)[0].value);
+      const userAction = decodeMessage(action.get(1)[0].value);
+      const userMessage = decodeMessage(userAction.get(1)[0].value);
+      const userText = Buffer.from(userMessage.get(1)[0].value).toString("utf8");
+      expect(userText).toContain("be brief");
+      expect(userText).toContain("hi");
     });
 
     it("encodes mcp_tools (field 4) when tools are provided", () => {