intel: don't use of intel_gpu_top

It's too much of a hassle to use intel_gpu_top, let's just wait
until these values are exposed in sysfs and until then we will just
have gpu load

src/intel.cpp

@@ -99,7 +99,6 @@ FILE* Intel::find_fd() {
                 if (found_driver){
                     if(strstr(line, "drm-engine-render")){
                         sscanf(line, "drm-engine-render: %" SCNu64 " ns", &val);
-                        if (val > 0)
                             return file;
                     }
                 }

src/intel.h

@@ -32,7 +32,7 @@ class Intel {
                 runtime = true;
 
             fdinfo = find_fd();
-            thread = std::thread(&Intel::intel_gpu_thread, this);
+            // thread = std::thread(&Intel::intel_gpu_thread, this);
         }
 
         void update() {
@@ -42,10 +42,10 @@ class Intel {
             gpu_info = gpu_info_intel;
         }
 
-        ~Intel(){
-            stop = true;
-            thread.join();
-        }
+        // ~Intel(){
+        //     stop = true;
+        //     thread.join();
+        // }
 };
 
 extern std::unique_ptr<Intel> intel;