Add RELEASE.md: pgxntool here creates branch, not tag

[jnasbyupgrade]

Summary

• Adds a minimal local RELEASE.md documenting one genuine, repo-specific deviation from the org's new shared release process doc (Add shared RELEASE.md for PGXN-extension repos ai#8, not yet merged): this repo's vendored pgxntool predates git-tag-based releases, so make tag/make dist create and push a git branch named after the version, not a git tag.

Why this is real, not stale docs

pgxntool/base.mk's tag target here is:

tag:
	git branch $(PGXNVERSION)
	git push --set-upstream origin $(PGXNVERSION)

git tag -l in this repo is empty. Instead there are branches 0.1, 0.1.0, 0.1.1, 0.2.0 — verified to each be an ancestor of master (real historical release points produced by make tag, not stale forks or abandoned work). pgxntool/README.asc in this repo's vendored copy also still says "make tag will create a git branch," consistent with the code.

The shared ../ai/RELEASE.md doc actually anticipates this exact situation in its gotchas section ("check the vendored version's actual behavior if a repo's docs still say 'branch'") — but assumes tags by default everywhere else (safety-check step, tag/dist step, idempotency behavior). Without a local note, following it here verbatim would be misleading in a few concrete ways spelled out in the new file (comparing against "the tag" when there is none, and different idempotency: git branch errors on re-creation even when it would point at the same commit, unlike a tag).

Dependency

This PR references ../ai/RELEASE.md, which is not yet merged — see Postgres-Extensions/ai#8. Marking this draft until that lands; the content here stands on its own either way, but the link target won't resolve until then.

Test plan

• Confirmed git tag -l is empty and 0.1/0.1.0/0.1.1/0.2.0 are branches, each an ancestor of master (git merge-base --is-ancestor <branch> master).
• Confirmed pgxntool/base.mk's tag/dist/rmtag/forcetag targets operate on branches (git branch, not git tag).
• Confirmed pgxntool/README.asc (vendored copy) still documents "branch" behavior.
• No code/build changes; docs-only.

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Organization UI

Review profile: ASSERTIVE

Plan: Pro Plus

Run ID: 2a5541c2-ca9d-469a-a8d0-5bb4b1abc725

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}