feat: Add strip_p filter to remove p tag around text

[pawamoy]

Sometimes we convert a single line of Markdown and we don't want to keep the surrounding <p> tags. This filter allows to strip them.

Example use-case:

<details class="{{ section.value.kind }}">
  <summary>{{ section.title|convert_markdown(heading_level, html_id)|strip_p }}</summary>
  {{ section.value.contents|convert_markdown(heading_level, html_id) }}
</details>

The filter does nothing if the markup is not wrapped in <p>.

[oprypin]

Wait, that's not so nice to do 😅
I'll try if there's a nicer alternative

[oprypin]

I have provided my attempt in #369. It's comparatively heavy, though :/

Which things I thought might be downsides and that I managed to resolve:

• I feel like the convert_markdown filter itself should be removing the tag, with a parameter. The current approach implies that there isn't any way to not have that tag there in the first place. But why (other than implementation details) should it be a separate step?

• What if the content is several paragraphs but it still tries to remove them? <p>Foo</p><p>Bar</p>

• (Far-fetched but) what if the p tag has attributes? <p x="y">Bar</p>
  Or generally this approach could somehow remove the start tag but not the end tag.

[pawamoy]

But why (other than implementation details) should it be a separate step?

No reason at all, I tried adding the option to the convert_markdown filter and it was ugly 😅
Thanks again for your better PR 🙂