Issue with log transformation in `fancplot -p square`

[kalavattam]

Hi again,

I believe I have encountered a bug when using the -l flag for log transformation in fancplot -p square.

I am using version 0.9.27 of FAN-C:

❯ fancplot --version
0.9.27

When I attempt to plot with the -l flag, I encounter the following error:

❯ fancplot \
>     -o "${out}" \
>     "${coord}" \
>     -p ${shape} -l -vmin ${vmin} -vmax ${vmax} \
>     --title "${title}" \
>     -c "${matcol}" \
>     "${i}"
2023-10-12 10:33:15,173 INFO Using non-interactive backend
2023-10-12 10:33:15,573 INFO Found 1 regions
/home/kalavatt/miniconda3/envs/fanc_pip_env/bin/fancplot:187: UserWarning: There was an error with plot 0, region XII:1-1078177 (Traceback (most recent call last):
  File "/home/kalavatt/miniconda3/envs/fanc_pip_env/bin/fancplot", line 170, in __init__
    fig, axes = gf.plot(plot_region)
                ^^^^^^^^^^^^^^^^^^^^
  File "/home/kalavatt/miniconda3/envs/fanc_pip_env/lib/python3.11/site-packages/fanc/plotting/plotter.py", line 211, in plot
    p.plot(r)
  File "/home/kalavatt/miniconda3/envs/fanc_pip_env/lib/python3.11/site-packages/fanc/plotting/base_plotter.py", line 785, in plot
    plot_output = self._plot((x_region, y_region))
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/kalavatt/miniconda3/envs/fanc_pip_env/lib/python3.11/site-packages/fanc/plotting/hic_plotter.py", line 299, in _plot
    self._refresh(region)
  File "/home/kalavatt/miniconda3/envs/fanc_pip_env/lib/python3.11/site-packages/fanc/plotting/hic_plotter.py", line 355, in _refresh
    self.im = self.ax.imshow(color_matrix, interpolation='none', aspect='auto',
              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/kalavatt/miniconda3/envs/fanc_pip_env/lib/python3.11/site-packages/matplotlib/__init__.py", line 1465, in inner
    return func(ax, *map(sanitize_sequence, args), **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/kalavatt/miniconda3/envs/fanc_pip_env/lib/python3.11/site-packages/matplotlib/axes/_axes.py", line 5756, in imshow
    im._scale_norm(norm, vmin, vmax)
  File "/home/kalavatt/miniconda3/envs/fanc_pip_env/lib/python3.11/site-packages/matplotlib/cm.py", line 436, in _scale_norm
    self.autoscale_None()
  File "/home/kalavatt/miniconda3/envs/fanc_pip_env/lib/python3.11/site-packages/matplotlib/cm.py", line 673, in autoscale_None
    self.norm.autoscale_None(self._A)
  File "/home/kalavatt/miniconda3/envs/fanc_pip_env/lib/python3.11/site-packages/matplotlib/colors.py", line 1771, in autoscale_None
    in_trf_domain = np.extract(np.isfinite(self._trf.transform(A)), A)
                                           ^^^^^^^^^^^^^^^^^^^^^^
  File "/home/kalavatt/miniconda3/envs/fanc_pip_env/lib/python3.11/site-packages/matplotlib/transforms.py", line 1505, in transform
    raise ValueError(
ValueError: Input values must have shape (N, 1) or (1,)
)
  warnings.warn("There was an error with plot {}, region {}:{}-{} ({})".format(i, region.chromosome,
2023-10-12 10:33:15,790 WARNING Input values must have shape (N, 1) or (1,)
Closing remaining open files:10_fanc_XII_KR-filt-0.2/MC-2019_Q_WT_repM.standard-rDNA-complete.mapped.6400.downsample-to-G1.hic...done

However, when I plot without the -l flag, everything works as expected:

❯ fancplot \
>     -o "${out}" \
>     "${coord}" \
>     -p ${shape} -vmin ${vmin} -vmax ${vmax} \
>     --title "${title}" \
>     -c "${matcol}" \
>     "${i}"
2023-10-12 10:33:41,264 INFO Using non-interactive backend
2023-10-12 10:33:41,579 INFO Found 1 regions
Closing remaining open files:10_fanc_XII_KR-filt-0.2/MC-2019_Q_WT_repM.standard-rDNA-complete.mapped.6400.downsample-to-G1.hic...done

Also, fancplot -p triangular works as expected with or without the -l flag, so it seems to be something to do with square. Have not tested split or mirror.

Thank you.