Run Prettier on the whole project

[lmac-1]

Short Description

Runs yarn prettier --write . over the whole project, plus three small fixes so nothing renders differently. Based on #869: merge that first, then retarget this to main.

Part of #866.

Details

Review commit by commit:

1. Add blank line after two :::tip openers. Without it, Prettier pulls the text onto the :::tip line and Docusaurus shows it as the title (api-discovery.md, migration-steps.md).
2. Run Prettier on the whole project. Formatting only, 151 files, nothing else. Skim it.
3. Put two admonition closers back on their own line. The reformat pulled the closing ::: in security-for-devs.md onto the paragraph above. collaboration.md was already broken the same way on main.
4. Ignore the reformat in git blame. Adds .git-blame-ignore-revs with commit 2's SHA. GitHub reads it automatically. Locally, run git config blame.ignoreRevsFile .git-blame-ignore-revs.

⚠️ Merge with "Create a merge commit" only, not squash or rebase. Both squash and "Rebase and merge" create new commits, so the SHA in .git-blame-ignore-revs would point at a commit that isn't on main. If this branch gets rebased (for example onto main after #869 lands), update the SHA to the new "Run Prettier" commit before merging.

Merge order: #874 rewrites links in 35 of these files, so the two will conflict. Merge #874 first. Then, rather than resolving conflicts by hand, merge main into this branch and re-run Prettier on the conflicting files. That re-run is a new formatting commit, so add its SHA to .git-blame-ignore-revs as well.

adaptors/library/, versioned_docs/ and static/ are in .prettierignore (#869), so they aren't touched. i18n/ is already formatted in #848.

Checked: yarn prettier --check . passes. yarn generate-library && yarn generate-adaptors && yarn build passes, with broken-anchor warnings the old PR (#870) reported as already present. No new :::tip Some text lines, and no ::: closers joined onto text.

AI Usage

Please disclose how you've used AI in this work (it's cool, we just want to
know!):

• I have used Claude Code
• I have used another model
• I have not used AI

You can read more details in our
Responsible AI Policy

[lmac-1]

This will be rerun after some other PRs have merged. Marking as draft for now.