Schur leads with additional self-energies are broken

[pablosanjose]

MWE:

julia> glead = LP.honeycomb() |> hopping(1) |> supercell((1,-1), region = r -> 0<=r[2]<=5) |> attach(nothing, cells = SA[5]) |> greenfunction(GS.Schur(boundary = 0));

julia> g = LP.honeycomb() |> hopping(1) |> supercell(region = r -> -6<=r[1]<=6 && 0<=r[2]<=5) |> attach(glead, region = r -> r[1] > 5.1) |> greenfunction
ERROR: ArgumentError: The contact surface has 6 sites, which doesn't match the 12 sites in the lead surface

The problem is SelfEnergy(hparent::AbstractHamiltonian, glead::GreenFunctionSchurLead; reverse = false, transform = missing, sites...), which builds lslead as a full unit cell, whereas lsparent just matches the surface of the lead unitcell