deseas issues at boundaries

[aekiss]

There are a couple of bugs in topography_number_seas. These aren't worth fixing as they don't seem to cause problems, but I'm noting them here for reference in case it's useful later. They could both be fixed by rewriting topography_number_seas to use an enlarged seas array with copies of cells from the other side of the left, right and top boundaries, like depth_halo in topography_nonadvective.

1. C-grid connectivity rules are used at the left and right boundaries, e.g. here, even if --grid_type B is specified
2. the top boundary is not joined up like a tripole seam - e.g. the little boxes at the top of test_topo.png are converted to land, even though they connect to the ocean via the tripole seam

test_topo.png (white=ocean):

test_topoC.nc from topogtools deseas --input test_topo.nc --output test_topoC.nc --grid_type C (red=ocean):