Manage public keys, over Stalwart's x:PublicKey registry

A new Settings section, next to Identities & signatures: list, add,
rename and remove the OpenPGP public keys and S/MIME certificates
published on this account. Only public material -- no private key is
stored, requested or sent by any of this.

This is PR #67 revived. That branch was built against 0.16.19, closed
unmerged on 2026-08-26, and shares no ancestry with main after the email
scrub, so it is ported rather than rebased: the four files it added are
carried over, the three it edited are applied by hand, and everything it
claimed was re-probed against the live 0.16.20 on 2026-09-05. The i18n
work is new -- nine catalogues landed on 2026-08-31, after that branch
was written.

What the re-probe confirmed, unchanged from 0.16.19:

  - An ordinary user may read *and* write their own keys, though the
    permissions table lists every sysPublicKey* permission as
    administrative. get and query both answered for a normal account,
    and a malformed create came back invalidProperties naming `key`
    rather than forbidden -- a rejection of the key, not of the person.

  - The server parses the key and says precisely what is wrong. So
    ihasmail does not validate key material; the server's message is
    shown verbatim, as password-policy rejections already are.

  - urn:stalwart:jmap is still absent from the session's top-level
    capabilities and present per-account, so the check that reads all
    three places is still the one that works.

What it added, none of which was known before:

  - A key can parse perfectly and still be refused, with different
    words: a sign-and-certify key with no encryption subkey -- what
    `gpg --quick-generate-key` produces -- gets "Could not find any
    suitable keys in OpenPGP public key". That is the rejection somebody
    exporting from GnuPG will actually meet, and it is not a paste
    error, so collapsing both to "invalid key" would send them back to
    the clipboard for a problem that is in the key.

  - emailAddresses comes back as {} when empty -- an object where a JMAP
    list property should be an array. It type-checks, then throws in
    join() while the list renders. normalize() checked the shape
    already; there is now a test saying why, and the mock answers {} the
    same way, because one that helpfully returned [] would let that
    crash ship.

  - A create answers with the id alone, no createdAt, so adding a key
    reloads rather than believing the response.

  - destroy works and leaves the registry empty. PR #67 shipped that
    path untested -- its live probe was refused before anything was
    created, so there was nothing to destroy.

  - Patching `key` is allowed by the server. The mock still refuses it,
    now deliberately rather than for want of evidence: ihasmail replaces
    a key by adding one and removing the old, which keeps createdAt
    meaning what it says.

x:EncryptionAtRest still does not exist on 0.16.20 -- asking for it is
an unknownMethod. encryptionAtRest is a field on x:AccountSettings, and
its value is a typed object ({"@type":"Disabled"}) rather than the bare
string ROADMAP described. Nothing here writes it.

An empty description is now sent as empty rather than filled in with
"Key". The description is stored on the server, so a default invented in
the client would be whichever language the adder happened to be using;
the list labels a blank one at render time instead.

Verified in a browser against the mock, not only in tests: both
rejections reach the toast in the server's own words with the form still
filled in, a good key renders its card, the kind is labelled from the
armour header, renaming persists, removing asks first and empties the
list, and the whole section reads correctly in German.

FEATURES.md

@@ -1212,6 +1212,64 @@ Over Stalwart's own registry objects, so there is no administrator in the loop:
   credentials". Doing it properly means implementing OAuth; that is in
   [ROADMAP.md](ROADMAP.md).
 
+## Encryption keys
+
+Settings › Encryption keys publishes OpenPGP public keys and S/MIME certificates
+on the account, over Stalwart's `x:PublicKey` registry: list, add, rename,
+remove. Only public material — no private key is stored, requested or sent by
+any of it.
+
+Everything below was established against a live Stalwart **0.16.20** on
+2026-09-05, rather than taken from the documentation, which is wrong on the
+first two:
+
+- **An ordinary user may read *and* write their own keys.** Stalwart's
+  permissions table lists every `sysPublicKey*` permission as administrative.
+  The server granted them to a normal account: `x:PublicKey/get` and
+  `x:PublicKey/query` both answered, and a create carrying a deliberately
+  malformed key came back `invalidProperties` naming `key` — a rejection of the
+  key, not of the person. Had the documentation been right this section would
+  have been useless to everyone but an administrator.
+- **Stalwart parses the key itself** and says precisely what is wrong with it:
+  *"Failed to decode OpenPGP public key: Malformed packet: Malformed CTB: MSB of
+  ptag not set."* So ihasmail does not validate key material. Anything it
+  checked would be a second opinion, and the one that counts would still be the
+  server's — its message is shown verbatim, the way password-policy rejections
+  already are.
+- **A key that parses can still be refused, and says something else.** A
+  sign-and-certify key with no encryption subkey — which is what
+  `gpg --quick-generate-key` produces by default — comes back *"Could not find
+  any suitable keys in OpenPGP public key"*. That is the rejection somebody
+  exporting from GnuPG is most likely to meet, and it is not a paste error:
+  the fix is to add an encryption subkey, not to paste more carefully. Showing
+  the server's own two sentences distinguishes the two; showing "invalid key"
+  would not.
+- **`emailAddresses` arrives as `{}` when it is empty** — an object, where a
+  JMAP list property should be an array. ihasmail checks the shape rather than
+  trusting the type, so nothing puts an object through `join()` halfway through
+  rendering the list.
+- **Creating a key answers with its id and nothing else**, so adding one
+  reloads the list instead of believing what came back. Renaming and removing
+  both work; removal leaves the registry genuinely empty.
+
+The kind badge — OpenPGP or S/MIME — is read from the armour header alone, which
+is a label rather than a parse, and the excerpt beside each key is deliberately
+not called a fingerprint: computing a real one means parsing the key, and naming
+it a fingerprint would invite somebody to verify against it.
+
+**What this does not do.** Nothing signs, encrypts, decrypts or verifies with
+these keys. The page says so. Adding one does not by itself start encrypting
+your mail — the one thing known to consume a registered key is
+`encryptionAtRest`, a field on `x:AccountSettings` beside `description`,
+`locale` and `timeZone`. There is no `x:EncryptionAtRest` object whatever the
+documentation suggests — asking for one on 0.16.20 is an `unknownMethod` — and
+the field's value is a typed object, `{"@type": "Disabled"}`, not a bare string.
+ihasmail does not yet offer it, and it carries a caveat of its own: turning it
+off does not decrypt what is already there. Signing and decrypting need a private key in a page
+served by the same host that would handle it, which is a security model to agree
+before it is a feature to build. Verifying a signature needs only public keys
+and is tractable on its own. See [ROADMAP.md](ROADMAP.md).
+
 ## Privacy by default
 
 Remote images blocked, the proxy on, read receipts never automatic, no