Extractor for rheinmaintv.de.

Add new extractor to _extractors.py.

Fix a potential problem in _html_search_regex: _search_regex may return a tuple when several subpatterns are selected. (Moreover, the result of clean_html is stripped already.)

Fixed the test cases.

Cosmetic chenges.

Fixed indendation. Oops!

Cosmetic changes.

Test cases completed.

Changed comments to make linter (flake8) happy. (Although commented-out code should in fact start with ##.)

Cosmetic changes

Linter (flake8) did not approve the line splits...

Improved video_id and JSON-LD extraction.

Improved/fixed fallbacks.

Use library function instead of quick&dirty solution.
(Plus, yet another small change of layout.)

Oops, still overlooked one issue... Use library function instead of new Python method. (Although this part of the code should, hopefully, never be executed.)

Further improved _html_search_regex.

Leave the formats alone. The final extension should be established elsewhere.

If the info_dict contains an extension, save it as the preferred format for merge (if not specified otherwise). In case of a merge, the extension will be overridden by the chosen (best) format without notice (for backwards compatibility, as the comment says).

This looks very ugly but still safer than overriding extensions in all available formats. It would probably be much better to leave the info_dict alone.

Revert the change.

Reverted the change and sorted the imports.

Fixed a glitch during the merge.

Consider any format (file extension) in the info_dict when merging video/audio formats.

Added another test and improved a comment.

Undo the latest changes.

old `_VALID_URL` was not matching URLs w/ alphanumeric `display_id`s like https://www.rheinmaintv.de/sendungen/beitrag-video/bricks4kids/vom-22.06.2022/

Co-authored-by: pukkandan <pukkandan.ytdlp@gmail.com>