search: email address search + address-first ranking; inline images; folder i18n; link security

Search (PG backend):
- '@' terms match address substrings (from/to/cc), so a partial email like
  "dyrektor@wfdif" resolves and narrows predictably as you type
- index each address split into parts, so a bare domain/local word ("wfdif")
  finds the address (indexed as one lexeme otherwise); dotted terms fall back to
  an address substring
- rank sender/recipient name+address matches above body/subject-only ones
  (addr_hit tier, then recency); cursor extended to keep keyset pagination correct
- scripts/reindex_address_parts.py backfills address tokens into existing rows

Inline images:
- extract attachments via a full MIME walk instead of shallow iter_attachments(),
  which missed inline images nested in mixed>related>alternative (cid: rendered
  broken); has_attachments counts only real (non-cid) attachments
- scripts/backfill_inline_attachments.py stores previously-missed inline images

Folder-name i18n (SPEC §12 M2):
- accent/punctuation-insensitive, multi-delimiter (/ \ .) special-use inference
  across ~13 languages; curated names (no substring false positives)

Link security (reading panel):
- links never navigate the message iframe: a parent-side click interceptor opens
  http/https in a new tab (noopener) and mailto in the composer (recipient
  pre-filled); all other schemes blocked

Backend 292 tests green, ruff + mypy --strict clean; frontend tsc/eslint/vitest clean.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01U72FZiF1W1pb8XxyWBTREA

backend/app/sync/folder_names.py

@@ -1,48 +1,120 @@
-"""Special-use inference from folder names (SPEC §6.3 fallback heuristics).
+"""Special-use inference from folder names (SPEC §6.3 / §12 M2 fallback).
 
-Used only when the server does not advertise SPECIAL-USE. Includes Polish names
-(Wysłane, Kosz, ...) alongside English/German, per the spec's i18n intent.
+Used only when the server does not advertise the SPECIAL-USE extension. Matching
+is accent- and punctuation-insensitive (so "Wysłane", "Éléments envoyés" and
+"Junk E-mail" all resolve), covers the hierarchy leaf under any of the common
+IMAP delimiters (``/  \\  .``), and recognises provider prefixes like
+``[Gmail]/`` implicitly via the leaf. Names are curated per language rather than
+substring-matched, so a folder like "Sent to John" is *not* misfiled as Sent.
 """
 
 from __future__ import annotations
 
+import unicodedata
+
+# Latin letters NFKD doesn't decompose to ASCII; fold them explicitly.
+_LATIN_SPECIAL = str.maketrans(
+    {"ł": "l", "ø": "o", "đ": "d", "ð": "d", "ß": "ss", "æ": "ae", "œ": "oe", "ı": "i", "þ": "th"}
+)
+
+
+def _compact(name: str) -> str:
+    """Fold to a comparable key: lowercase, strip accents/diacritics, drop every
+    non-alphanumeric char. "Éléments envoyés" -> "elementsenvoyes"."""
+    folded = unicodedata.normalize("NFKD", name.strip().lower()).translate(_LATIN_SPECIAL)
+    return "".join(ch for ch in folded if ch.isalnum())
+
+
+# Localized folder names by special_use. Written readably; compacted at load.
+_NAMES: dict[str, list[str]] = {
+    "sent": [
+        "sent", "sent items", "sent mail", "sent messages",  # English / Outlook / Gmail
+        "wysłane", "elementy wysłane", "wiadomości wysłane",  # Polish
+        "gesendet", "gesendete objekte", "gesendete elemente",  # German
+        "envoyés", "éléments envoyés", "messages envoyés", "courrier envoyé",  # French
+        "enviados", "elementos enviados", "correo enviado",  # Spanish
+        "inviata", "posta inviata", "messaggi inviati",  # Italian
+        "verzonden", "verzonden items", "verzonden berichten",  # Dutch
+        "enviadas", "itens enviados", "mensagens enviadas",  # Portuguese
+        "отправленные", "исходящие",  # Russian
+        "skickat", "skickade", "skickade objekt",  # Swedish
+        "sendt", "sendte", "sendte elementer",  # Norwegian / Danish
+        "lähetetyt",  # Finnish
+        "odeslané", "odeslaná pošta",  # Czech
+    ],
+    "drafts": [
+        "drafts", "draft",
+        "robocze", "kopie robocze", "wersje robocze",
+        "entwürfe", "entwurf",
+        "brouillons",
+        "borradores",
+        "bozze",
+        "concepten",
+        "rascunhos",
+        "черновики",
+        "utkast",
+        "kladder",  # Danish
+        "luonnokset",  # Finnish
+        "koncepty",  # Czech
+    ],
+    "trash": [
+        "trash", "deleted", "deleted items", "deleted messages", "bin", "recycle bin",
+        "kosz",
+        "papierkorb", "gelöschte objekte", "gelöschte elemente",
+        "corbeille", "éléments supprimés",
+        "papelera", "elementos eliminados",
+        "cestino", "posta eliminata",
+        "prullenbak", "verwijderde items",
+        "lixeira", "lixo", "itens excluídos",
+        "корзина", "удаленные",
+        "papperskorg", "borttaget",
+        "papirkurv", "slettet",
+        "roskakori", "poistetut",
+        "koš",  # Czech
+    ],
+    "junk": [
+        "junk", "junk email", "junk e-mail", "spam", "bulk mail", "bulk",
+        "wiadomości-śmieci", "niechciane",
+        "unerwünscht", "werbung",
+        "courrier indésirable", "indésirables", "pourriel",
+        "correo no deseado", "no deseado", "correo basura",
+        "posta indesiderata", "indesiderata",
+        "ongewenst", "ongewenste e-mail",
+        "lixo eletrônico",
+        "спам", "нежелательная почта",
+        "skräp", "skräppost",
+        "uønsket",
+        "roskaposti",
+        "nevyžádaná pošta",
+    ],
+    "archive": [
+        "archive", "archives", "all mail",
+        "archiwum",
+        "archiv",
+        "archivio",
+        "archivo",
+        "archief",
+        "arquivo", "arquivo morto",
+        "архив", "вся почта",
+        "arkiv",
+        "arkisto",
+    ],
+}
+
 _NAME_MAP: dict[str, str] = {
-    # sent
-    "sent": "sent",
-    "sent items": "sent",
-    "sent mail": "sent",
-    "wysłane": "sent",
-    "wyslane": "sent",
-    "elementy wysłane": "sent",
-    "gesendet": "sent",
-    # drafts
-    "drafts": "drafts",
-    "draft": "drafts",
-    "robocze": "drafts",
-    "kopie robocze": "drafts",
-    "entwürfe": "drafts",
-    # trash
-    "trash": "trash",
-    "deleted": "trash",
-    "deleted items": "trash",
-    "kosz": "trash",
-    "papierkorb": "trash",
-    # junk
-    "junk": "junk",
-    "spam": "junk",
-    "bulk mail": "junk",
-    "wiadomości-śmieci": "junk",
-    # archive
-    "archive": "archive",
-    "archiwum": "archive",
-    "all mail": "archive",
-    "archiv": "archive",
+    _compact(name): special for special, names in _NAMES.items() for name in names
 }
 
 
 def infer_special_use(path: str) -> str | None:
-    """Infer a special_use value from a mailbox path/name (case-insensitive)."""
-    if path.upper() == "INBOX":
+    """Infer a special_use value from a mailbox path/name (accent-insensitive)."""
+    if path.strip().upper().rstrip("/.") == "INBOX":
         return "inbox"
-    leaf = path.replace("\\", "/").split("/")[-1].strip().lower()
-    return _NAME_MAP.get(leaf) or _NAME_MAP.get(path.strip().lower())
+    # Take the hierarchy leaf under any common IMAP delimiter (/ \ or . for
+    # Dovecot/Courier namespaces), then fold for lookup.
+    normalized = path
+    for delim in ("/", "\\", "."):
+        normalized = normalized.replace(delim, "\n")
+    segments = [s for s in normalized.split("\n") if s.strip()]
+    tail = segments[-1] if segments else path
+    return _NAME_MAP.get(_compact(tail)) or _NAME_MAP.get(_compact(path))