Merge pull request #1358 from sacereda/optimizeMeshes

Low performance in OptimizeMeshesProcess with huge number of meshes

code/OptimizeMeshes.cpp

@@ -181,11 +181,8 @@ void OptimizeMeshesProcess::ProcessNode( aiNode* pNode)
                     verts += mScene->mMeshes[am]->mNumVertices;
                     faces += mScene->mMeshes[am]->mNumFaces;
 
+                    pNode->mMeshes[a] = pNode->mMeshes[pNode->mNumMeshes - 1];
                     --pNode->mNumMeshes;
-                    for( unsigned int n = a; n < pNode->mNumMeshes; ++n ) {
-                        pNode->mMeshes[ n ] = pNode->mMeshes[ n + 1 ];
-                    }
-
                     --a;
                 }
             }