Combine single and tricolor echogram plotting

[leewujung]

In the main branch, the code for single echogram and tricolor echogram plotting are separated. @shivam096 has previously done some work (which lives in the refactor_main branch) but that has not been merged into main. This issue to ensure that the existing code in refactor_main is revived (since it's been almost a year) and merged into main, with demo notebooks in the docs updated with the new function calls.

[leewujung]

@iamluan: this should be a relatively small task, but would get you familiar with the current logic in the codebase.

cc @LOCEANlloydizard for visibility.

[iamluan]

I've reviewed PR #193, which addresses the issue you mentioned. Since main and refactor_main have diverged significantly since then, I think we might create a new PR based on the original PR to integrate the creation of the single echogram and tricolor echogram.
I have tested the echogram example with the code in refactor_main and encountered an error with colormaps, as shown below.

raise ValueError(f"Supplied cmap {cmap} not found among {providers_checked} colormaps.")
ValueError: Supplied cmap #FFFFFF not found among matplotlib, bokeh, or colorcet colormaps.

The code that produces this error is from the notebook in the main docs/source/version_0.1.0/echogram_examples.ipynb

eg = MVBS_ds.eshader.echogram2(
    channel = [
        'GPT  18 kHz 009072058c8d 1-1 ES18-11',
        'GPT  38 kHz 009072058146 2-1 ES38B',
        'GPT 120 kHz 00907205a6d0 4-1 ES120-7C',
        ],
    cmap = [
        "#FFFFFF", "#9F9F9F", "#5F5F5F", "#0000FF", "#00007F", "#00BF00", "#007F00", 
        "#FFFF00", "#FF7F00", "#FF00BF", "#FF0000", "#A6533C",  "#783C28",
    ], 
    vmin = -80, 
    vmax = -30,
)
panel.Row(eg)

I will investigate the error further. If you have any thoughts on this, please let me know. Thanks @leewujung

[leewujung]

@iamluan : Hmm, interesting -- please take a look at this colormap code in echopype, which should be what's used here. Seems better if that's in a single place. Since echopype is already a requirement, seems straightforward to just use this new colormap from there.

Within this current issue I think we can just use what's in echopype. However, beyond this particular colormap (ek500) the considerations are a bit different. I just added some thoughts in the new issue #231 for new colormaps, and pinging mentor @LOCEANlloydizard there since we had some prior discussions about it. Feel free to contribute your thoughts there! Thanks!

[LOCEANlloydizard]

Hi both, indeed we should use the colormap in echopype. There is an example of its use here if that helps, @iamluan. Answering in #231 to discuss the broader scope! Cheers!

[iamluan]

Continue to the ValueError: Supplied cmap #FFFFFF not found among matplotlib, bokeh, or colorcet colormaps.. This happens because of line 114 in the code below: opts_dict["cmap"] has the value #FFFFFF; it should be a list of hex color strings as input. To fix this issue, we can add the following line within the if-else block at the beginning of the function: cmaps = [cmap]*len(channels).

https://github.com/OSOceanAcoustics/echoshader/blob/90992a362b2d135c16c49bfb0112099e172c83ed/echoshader/plots/echogram.py#L112-L115