Don't touch the pinning of the main thread when not using affinity.

There is a problem when the cpu mask is empty and using multiple MPI ranks on a node, in that case all the main threads end up pinned to one core during the reading of the initial condition (and any other tasks done before engine_init is called).