DOC: Add plotting examples and fix broken examples

[deppen8]

This PR improves the pandas.plotting._misc.py module docs by:

1. Adding example plots where they were missing from docstrings.
2. Fixing examples that were not formatted to render with

.. plot::
    :context: close-figs

3. Simplifying the syntax of some oddly-written examples.

In some cases, I replicated the examples found in the User Guide.

[pep8speaks]

Hello @deppen8! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found:

There are currently no PEP 8 issues detected in this Pull Request. Cheers! 🍻

Comment last updated at 2020-05-05 21:05:27 UTC

[datapythonista]

Thanks @deppen8, cool stuff. Did you render the website to see that everything is looking as expected?

@jorisvandenbossche, may be you want to have a look here.

[jorisvandenbossche]

Thanks! Looks good generally, just wondering: are the random seeds necessary? We don't really do this in other parts of docs, I think.

[jorisvandenbossche]

is this seed needed?

[deppen8]

@datapythonista, I built the docs locally and nothing looked out of order.

@jorisvandenbossche I found this advice which suggested using random seeds.

https://pandas.pydata.org/docs/development/contributing_docstring.html#tips-for-getting-your-examples-pass-the-doctests

In the case of the lag plot, it has an added benefit of keeping the demo data above zero, which probably matches most use cases better.

[jorisvandenbossche]

I found this advice which suggested using random seeds.

Ah, yes, but that's only to get doctests passing where eg a number is shown in the output, while here the plots are not actually checked.

For the lag plot that sounds like a good reason, for the others I would maybe remove it again, if we don't do that elsewhere

[deppen8]

For the lag plot that sounds like a good reason, for the others I would maybe remove it again, if we don't do that elsewhere

Done!

[jorisvandenbossche]

Thanks @deppen8 !