Add options `--forcegpu` and `--profile`

This PR:

• Restore the --forcegpu option to enforce kernels to run on the GPU whenever a CUDA implementation is available.
• Restore the --profile option to display the performance profile of each kernel and their distribution among the resources.
• Add a cmake option CHAMELEON_SIMULATION_EXTENDED to enable non GPU kernels to be easily simulated on the GPU through simgrid.

control/context.c

@@ -147,9 +147,11 @@ int CHAMELEON_Enable(int option)
             chamctxt->autotuning_enabled = CHAMELEON_TRUE;
             break;
         case CHAMELEON_PROFILING_MODE:
-            chamctxt->profiling_enabled = CHAMELEON_TRUE;
             RUNTIME_start_profiling();
             break;
+        case CHAMELEON_KERNELPROFILE_MODE:
+            chamctxt->profiling_enabled = CHAMELEON_TRUE;
+            break;
         case CHAMELEON_PROGRESS:
             chamctxt->progress_enabled = CHAMELEON_TRUE;
             break;
@@ -218,9 +220,11 @@ int CHAMELEON_Disable(int option)
             chamctxt->autotuning_enabled = CHAMELEON_FALSE;
             break;
         case CHAMELEON_PROFILING_MODE:
-            chamctxt->profiling_enabled = CHAMELEON_FALSE;
             RUNTIME_stop_profiling();
             break;
+        case CHAMELEON_KERNELPROFILE_MODE:
+            chamctxt->profiling_enabled = CHAMELEON_FALSE;
+            break;
         case CHAMELEON_PROGRESS:
             chamctxt->progress_enabled = CHAMELEON_FALSE;
             break;