Fix to #363. Drifting decision needs to use the same multipole information as the actual tree walk. Also fix some gravity_cache padding values.

Force compiler to report shadowed variables. Removed all shadowed variables. Added the function scheduler_activate_send() to simplify the unskipping of MPI sends.

When padding the gravity cache, use -2 * cell->width for the particle position to make sure this is an invalid (but reasonable) position.

More specialised timers for the gravity leaf-leaf interactions. Improvements to the vectorization of the PM leaf-leaf interaction.

Split the P2P and M2P into two separate (vectorizable) loops in the leaf-leaf gravity P-P interactions.

Correct use of caches in the self-PP gravity task. Make sure we can run without vectorization. Make sure we can run without hydro.

Correct use of caches in the self-PP gravity task. Make sure we can run without vectorization. Make sure we can run without hydro.

Do the P-P gravity calculations in the frame of the cells for more accuracy in the distance calculations.

runner_dopair_grav() should abort if the distance between the cells is larger than the mesh smoothing scale.

Split the gravity pp tasks into truncated and full Newtonian potential. Use the full potential when the particles are close enough.

Added a parameter to the gravity properties to contain the distance below which all computed forces are Newtonian

Make the long-range gravity task escape the cells that are beyond the range handled by the mesh forces.

runner_dopair_grav_pp() can now use the periodic boundary conditions. The L2L kernel is now always non-periodic as the shift should never cross a domain boundary.

Create the gravity pair tasks over the periodic boundary conditions. Make the long-range gravity task loop over the cells even over the periodic boundaries.