Throat cross-sectional area calculation is wrong when using high accuracy with real voxel sizes.

[Cesar-Timana]

Hello everyone,
There is an error in calculating the throat cross-sectional area using the high accuracy method.

The surface area used as input has to be without multiplying by the voxel size.

The following test demonstrates the problem:

import porespy as ps
import numpy as np
import openpnm as op
np.random.seed(13)

resolution = 5e-7

im = ps.generators.overlapping_spheres([100, 100, 100], r=7, porosity=0.7)

snow_output = ps.networks.snow2(im,
                   voxel_size=resolution,
                   boundary_width=3,
                   accuracy='high', # high
                   parallelization=None)

pn = op.io.network_from_porespy(snow_output.network)

A = pn['throat.cross_sectional_area']
print(A[1:10])

When subtracting the surface area of merged regions from the isolated region surface areas, they need to be in the same unit.