Perform MPI_ALLREDUCE operations in place when possible

[mathrack]

Several all-to-all MPI reductions are using a temporary variable :

Incompact3d/src/Case-TGV.f90

Line 298 in 5565268

call MPI_ALLREDUCE(temp1,enst,1,real_type,MPI_SUM,MPI_COMM_WORLD,code)

However, MPI reductions can be performed in place to avoid temporary variables :

Incompact3d/src/navier.f90

Line 1541 in 5565268

call MPI_ALLREDUCE(MPI_IN_PLACE,qm,1,real_type,MPI_SUM,MPI_COMM_WORLD,code)

It would be a nice improvement to perform MPI all-to-all reductions in place when possible and to remove temporary variables, especially when their name is not clear (temp1, temp2, ...)