WEB-2754 - format link fixes

[davidejones]

What does this PR do?

site-region can also have its own link scope.
As this wasn't being accounted for it would leak into the "main" section causing the format link to say there are duplicates

EDIT:
This PR:

• removes the precommit hook usage of the format link script
• removes the integration processing usage of the format link script
• Keeps the reworked format link script for debugging purposes

Motivation

https://datadoghq.atlassian.net/browse/WEB-2754

Preview

Testing
Locally i tested the following file

local/bin/py/build/actions/format_link.py -f content/en/continuous_integration/setup_pipelines/custom_tags_and_metrics.md

Trying on all files with something like

 for f in $(find ./content/en -type f -iname '*.md'); do local/bin/py/build/actions/format_link.py -f "$f"; done

However we should test multiple file changes and see how the content changes.

EDIT:
We should now just test that integrations are displaying as intended on the preview:

https://docs-staging.datadoghq.com/david.jones/format-link-fixes/integrations/

Additional Notes

---

Reviewer checklist

• Review the changed files.
• Review the URLs listed in the Preview section.
• Check images for PII
• Review any mentions of "Contact Datadog support" for internal support documentation.

[hestonhoffman]

Hi David, I ran this on a test file and tried to break it and it looks like it's working really well!

[hestonhoffman]

@davidejones I just tried this on a file I've been writing, and while it doesn't throw an error like the old script did (I think it's cos I'm using single line code-blocks), it doesn't renumber the references.

Here's the file for reference
high_memory_usage.md

[davidejones]

@hestonhoffman So generally i've tried to make the script avoid re-ordering where possible for 2 reasons.

• The diffs are easier to read on modified files if there are less changes
• The order Isn't particularly important for being able to render the page

I can change this though, would docs prefer the same behavior as before?

[hestonhoffman]

@davidejones Yeah, that makes sense. I prefer this behavior. It should prevent the issue where merging master into a feature bunch pulls a bunch of unrelated changes into the PR

[DavidWeid]

Does this need to be changed back for preview? @davidejones