Simplify return-format for dat-datasets

[haraldp271]

ImportRun presently returns a dat-dataset with a separate dictionary entry for each column:

Lev2=segment_utils.ImportRun(SRC, 2)
centerB = Lev2['Horizons']['AhB']['CoordCenterInertial']
print(centerB)

{'time': array([[0.0000e+00, 0.0000e+00],
        [5.0000e-01, 5.0000e-01],
        [1.0000e+00, 1.0000e+00],
        ...,
        [4.1385e+03, 4.1385e+03],
        [4.1390e+03, 4.1390e+03],
        [4.1395e+03, 4.1395e+03]]),
 'x': array([[ 0.00000000e+00, -3.97135358e+01],
        [ 5.00000000e-01, -3.97135217e+01],
        [ 1.00000000e+00, -3.97134437e+01],
        ...,
        [ 4.13850000e+03, -3.12849015e+01],
        [ 4.13900000e+03, -3.12940663e+01],
        [ 4.13950000e+03, -3.13031083e+01]]),
 'y': array([[ 0.00000000e+00,  2.08095432e-03],
        [ 5.00000000e-01, -3.84157027e-02],
        [ 1.00000000e+00, -7.86382932e-02],
        ...,
        [ 4.13850000e+03,  1.14077851e+01],
        [ 4.13900000e+03,  1.13613565e+01],
        [ 4.13950000e+03,  1.13148841e+01]]),
 'z': array([[ 0.00000000e+00,  2.81790998e-08],
        [ 5.00000000e-01,  2.81702228e-08],
        [ 1.00000000e+00,  2.81619053e-08],
        ...,
        [ 4.13850000e+03, -3.30562543e-06],
        [ 4.13900000e+03, -3.30750271e-06],
        [ 4.13950000e+03, -3.30937792e-06]])}

This generality is needed if the number of columns in a dat-file change between segments. However, it comes with significant inconvenience for the many cases where where the column-count is known to be constant.

Switch the code, such that by default it returns simple formats (i.e., one 4-column numpy array for this example). And explicitly hardcode into segment_utils.ImportRun the one case where the generality is needed.