docs: README banner — development moved to postguard, crate is cryptify/

[rubenhensen]

Part of postguard#294 (moving cryptify's issues into postguard ahead of archiving this repo). Adds the same "development moved" banner already used on postguard-outlook-addon's README, adjusted for the fact that this repo's Docker publish pipeline hasn't cut over yet (postguard#293) — archiving is deliberately not claimed here.

[dobby-coder]

Thanks for the context, Ruben! I'm on it — reviewing this PR now to check the "development moved" banner change and make sure it's consistent with the postguard-outlook-addon version while correctly leaving out the archiving claim since #293 hasn't landed yet.

[dobby-coder]

Reviewed after the fact: this merged at 12:45 on 2026-08-07, before the review ran, so treat this as a post-merge sign-off rather than a gate.

The banner was accurate when it merged. One line has since gone stale, and not through any fault of this PR: #206 merged 75 minutes later (14:00), stripped 104 lines of Docker build/push out of .github/workflows/ci.yml, and did not touch README.md. README line 8 now states the opposite of ci.yml's own header comment. Details inline. It needs a one-line follow-up PR against README.md, which I have not opened.

What I checked and found correct:

• cryptify really is a workspace member in postguard's root Cargo.toml (members = ["pg-core", "pg-cli", "pg-pkg", "pg-ffi", "cryptify"]), sharing the single root Cargo.lock with pg-core/pg-pkg.
• "Open issues have moved there too" holds: gh issue list --repo encryption4all/cryptify --state open returns nothing.
• postguard#293 (the Docker cutover) is CLOSED; postguard#294 (transfer the issues, then archive) is OPEN. Not claiming archival here was the right call.
• Banner placement fits the standardized README format: logo, docs link, banner, then the role paragraph.

One nit in that same line: it carries the only em dash in the README, which the org's README style rules exclude. The rewrite below drops it.

[dobby-coder]

This line is stale on main now, and it is the one place a reader would look before touching the publish pipeline.

It says the Docker cutover has not happened yet, and it tells the reader not to remove the build/push steps. #206 overtook both (merged 2026-08-07 14:00, 75 minutes after this PR): it removed those steps from .github/workflows/ci.yml (6 insertions, 104 deletions) and left README.md untouched. postguard#293, the cutover task this line points at, is CLOSED.

ci.yml's header comment now reads:

Docker publishing moved to encryption4all/postguard, which builds this crate as a workspace member and pushes the same ghcr.io/encryption4all/cryptify image name (postguard#293). This workflow no longer builds or pushes any image.

So the README and the workflow it describes contradict each other. The direction of the error is what makes it worth a follow-up: a reader trusting the README will preserve build/push steps that are already gone, or hold off archiving to wait for a cutover that already landed.

Suggested replacement for a follow-up PR (GitHub will not apply a suggestion block on a merged PR, so this is plain text):

> This repository is kept for history. Docker publishing has moved to `postguard`, which now pushes `ghcr.io/encryption4all/cryptify` (postguard#293). This repository's CI only runs the Rust quality gate and release-plz. Archiving is tracked in postguard#294.

That also drops the em dash, per the org README style rules.

[dobby-coder]

Reviewed this — looks like it merged before I got to it, so I left a post-merge review instead of a gate.

The banner itself was accurate at merge time, but heads up: #206 landed 75 minutes later and ripped the Docker build/push steps out of ci.yml without touching the README, so line 8's "don't remove the build/push steps yet" note is now stale. I left a suggested one-line replacement inline on README.md:8 — worth a quick follow-up PR to fix. Everything else checked out: cryptify's still a genuine workspace member sharing postguard's root Cargo.lock, no open issues left here, and I left postguard#294's archival claim alone since that's still open.