Split schema and row processing

The recursive calls into `read` caused unnecessary slowness in the
connection phase. The actual first row should only be read when the data
is requested. This can possibly solve the false timeout errors caused by
slow sending of the first row of data.

connectors/cdc-connector/cdc_connector.h

@@ -145,6 +145,7 @@ private:
     bool do_auth();
     bool do_registration();
     bool read_row(std::string& dest);
+    bool read_schema();
     void process_schema(json_t* json);
     SRow process_row(json_t*);
     bool is_error();