Closes #8734
Part of #15014
Related nvaccess/nvda-misc-deps#30, #16002, #15950, #15939, #15981

Summary of the issue:
In order to migrate to Crowdin, we must convert txt2tags to markdown.
In order transition safely, beta will build docs from t2t to markdown to html.
Eventually the t2t will be removed and the markdown will become the source of truth.

Description of user facing changes
The user Guide no longer has numbered sections

Translators and documentation writers now use extended markdown syntax rather than txt2tags.

Description of development approach
The build system performs certain pre-processing and post-processing when converting t2t to HTML.
This equivalent system should be retained - i.e. the translator and documentation contribution experience should remain the same for markdown to HTML.
Additionally, when converting t2t to markdown, new processing rules had to be created.

There is no universal standard for custom anchors in markdown.
As such text2tags doesn't have default rules for this. To retain our custom anchors, I added rules for a common markdown extended syntax.
Similarly a shortcut is used for generating table of contents.
See nvaccess/nvda-misc-deps#30

Setting the language code is done using the user_docs folder name, and the direction of RTL is manually set for the 3 current languages that require it, Persian, Arabic, Hebrew.

Special Catalan processing for adding hreflang attributes is converted to a markdown extension syntax of {hreflang=en}

The key commands file is generated using a custom made markdown python extension.

Fixes #16063

Summary of the issue:
#15950 converted definition lists to tables, however the conversion could have been done better.
Both lines of the table start with header syntax || whereas it should only be the first header.

Tables were previously converted from definition lists to look like this

Default	Enabled
Options	Default (Enabled), Enabled, Disabled
Options should be listed before default logically.
The first column should be headers, not the first row.
Unfortunately there is no good support for column headers in markdown in our markdown parsing stack.
As such, this can be improved by just having no headers.

Description of user facing changes
Change all option definition tables to be formatted as follows.
The header row is hidden in CSS

Options	Default (Enabled), Enabled, Disabled
Default	Enabled
Description of development approach
The following regex was used:

find: \|\|(.*)\|(.*)\|\n\|\|(.*)\|(.*)\|
replace: || . {.hideHeaderRow} | . |\n|$3|$4|\n|$1|$2|
Testing strategy:

Fixes nvaccess#14240

Summary of the issue:
Definition lists are not exposed correctly in NVDA.
They are also not supported in markdown natively.
In order to convert text2tags to markdown, these must be converted to tables

Description of user facing changes
Definition lists are now tables in the user guide, which are exposed better by NVDA.

The user guide standards have been updated to encourage the following format for documenting settings.
The standards have been expanded to further document settings in a well structured manner.
This is to improve readability, creating a simple standard to learn more about a settings documentation.
This also allows for the future of automatically collating tables into a settings reference similar to key commands.

Options	Default (Enabled), Enabled, Disabled
Default	Enabled
Toggle command	nvda+shift+e
Option	Behaviour
Enabled	behaviour of enabled
Disabled	behaviour of enabled
Description of development approach
All t2t files had definition lists converted to markdown.
The following regex was used to perform the conversion.

To remove the definition list terminator :

Find: ((^:.+\n  .*\n)+):
replace: $1
To replace rows:

Find: ^: (.*)\n  (.*)\n
Replace: || $1 | $2 |\n

Closes nvaccess#8734
Part of nvaccess#15014
Related nvaccess/nvda-misc-deps#30, nvaccess#16002, nvaccess#15950, nvaccess#15939, nvaccess#15981

Summary of the issue:
In order to migrate to Crowdin, we must convert txt2tags to markdown.
In order transition safely, beta will build docs from t2t to markdown to html.
Eventually the t2t will be removed and the markdown will become the source of truth.

Description of user facing changes
The user Guide no longer has numbered sections

Translators and documentation writers now use extended markdown syntax rather than txt2tags.

Description of development approach
The build system performs certain pre-processing and post-processing when converting t2t to HTML.
This equivalent system should be retained - i.e. the translator and documentation contribution experience should remain the same for markdown to HTML.
Additionally, when converting t2t to markdown, new processing rules had to be created.

There is no universal standard for custom anchors in markdown.
As such text2tags doesn't have default rules for this. To retain our custom anchors, I added rules for a common markdown extended syntax.
Similarly a shortcut is used for generating table of contents.
See nvaccess/nvda-misc-deps#30

Setting the language code is done using the user_docs folder name, and the direction of RTL is manually set for the 3 current languages that require it, Persian, Arabic, Hebrew.

Special Catalan processing for adding hreflang attributes is converted to a markdown extension syntax of {hreflang=en}

The key commands file is generated using a custom made markdown python extension.

Fixes nvaccess#16063

Summary of the issue:
nvaccess#15950 converted definition lists to tables, however the conversion could have been done better.
Both lines of the table start with header syntax || whereas it should only be the first header.

Tables were previously converted from definition lists to look like this

Default	Enabled
Options	Default (Enabled), Enabled, Disabled
Options should be listed before default logically.
The first column should be headers, not the first row.
Unfortunately there is no good support for column headers in markdown in our markdown parsing stack.
As such, this can be improved by just having no headers.

Description of user facing changes
Change all option definition tables to be formatted as follows.
The header row is hidden in CSS

Options	Default (Enabled), Enabled, Disabled
Default	Enabled
Description of development approach
The following regex was used:

find: \|\|(.*)\|(.*)\|\n\|\|(.*)\|(.*)\|
replace: || . {.hideHeaderRow} | . |\n|$3|$4|\n|$1|$2|
Testing strategy: