feat: a teaser's card references the drop it announced and the piece's variants (4402, 4401)

The operator's problem: a Patreon teaser is a pointer, and its card showed the
censored crop plus a text link while the content it pointed at sat on another
card. The fix is a REFERENCE, not an absorption: "the nested items on the
unified post are a duplicate or reference of existing content". Nothing is
written. Discord posts keep their own rows, dates and places in the feed.

- post_unification: for each teaser with a linked association, the drop's
  images and text, plus its variant family: Discord images sharing the
  seed's gated LEADING working name, or a phash near-duplicate, within a
  window of the teaser. One hop only, oldest first.
- Measured on artist 8 before writing it: of 121 message pairs 2-60 days
  apart that share a gated token, 106 share the leading name and all read
  as real families. Of the 15 sharing only a trailing word, 14 are sibling
  pieces and one is a plain collision (`bottom`, 56 days). The family cap
  is 8, not the pairing cap of 6, because `tentacooler` and `0-k1` (6 posts
  each) are real families.
- The feed drops a linked drop's own card only within discord_link_fold_hours
  of its teaser (default 24): "only hidden from the post view they're posted
  the same day". Older referenced posts stay where they landed.
- post_association.linked_by records whether FC or a person made the link,
  so the card can say so. Undo is the existing dismiss.
- `image0` (gallery-dl's fallback name) becomes a stopword.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LVjrnpQjRgHdvq95rASoiR

backend/app/services/post_naming.py

@@ -112,6 +112,10 @@ _STOPWORDS = frozenset({
     # `{user[name]}` as it for ~1,600 files, so it is the single most common
     # "name" in the library and identifies nothing.
     "none",
+    # gallery-dl's fallback when a Discord attachment has no filename of its
+    # own. Measured on artist 8: four unrelated images across 1,974 days, and
+    # the one false family the leading-name rule admitted inside 60 days.
+    "image0",
 })
 
 # A bare year: still needed for the TEXT signal, where words and numbers are
@@ -170,19 +174,17 @@ def _strip_prefixes(stem: str) -> str:
     return _HASH_SUFFIX.sub("", stem)
 
 
-def working_name_tokens(path: str) -> set[str]:
-    """The identity-bearing tokens in one image's filename.
+def _ordered_tokens(path: str) -> list[str]:
+    """The identity-bearing tokens of one filename, in the order written.
 
-    Returns an EMPTY set for a name that carries no working title — a
-    screenshot, a bare number, a stopword. Empty means "no evidence", which the
-    caller must treat as silence rather than as a weak match; see the module
-    docstring for the false positive that rule exists for.
+    The one tokenizer both public readings share, so the set of names and the
+    leading name cannot disagree about what counts as a name.
     """
     stem = _strip_prefixes(PurePosixPath(path).stem)
     if _SCREENSHOT.match(stem.strip()):
-        return set()
+        return []
 
-    out: set[str] = set()
+    out: list[str] = []
     # Hyphens are kept INSIDE tokens — `0-k` is a real working name on the live
     # instance, and splitting on hyphen would reduce it to a single character
     # and then discard it for being too short.
@@ -192,10 +194,43 @@ def working_name_tokens(path: str) -> set[str]:
             continue
         if tok in _STOPWORDS or not _HAS_LETTER.search(tok):
             continue
-        out.add(tok)
+        if tok not in out:
+            out.append(tok)
     return out
 
 
+def working_name_tokens(path: str) -> set[str]:
+    """The identity-bearing tokens in one image's filename.
+
+    Returns an EMPTY set for a name that carries no working title — a
+    screenshot, a bare number, a stopword. Empty means "no evidence", which the
+    caller must treat as silence rather than as a weak match; see the module
+    docstring for the false positive that rule exists for.
+    """
+    return set(_ordered_tokens(path))
+
+
+def leading_name(path: str) -> str | None:
+    """The FIRST identity token of a filename — the piece, not its decoration.
+
+    Creators lead with what the piece is and trail with what this export of it
+    is: `Year_20k_wip1`, `not_sombra_21-cumpeen`, `Tentacooler_c_ins`. Content
+    words sit at the tail, and they span too FEW posts for any frequency cap
+    to catch — `nude`, `cum` and `top` are on three of artist 8's posts each.
+    Position is what separates them from a name.
+
+    Measured on artist 8, Discord messages 2-60 days apart sharing a gated
+    token: 121 pairs. The 106 sharing the leading name all read as one piece's
+    trickle; of the 15 sharing only a trailing word, 14 are sibling pieces
+    (`Bea_Machamp_Shiny_*` / `Bea_Machoke_Shiny_*`) and one is a plain
+    collision (`Undyne_insert_bottom_only-C` / `Lichgalclc_Lingerie_Bottom_21`).
+
+    None when the name carries no identity at all — see `working_name_tokens`.
+    """
+    tokens = _ordered_tokens(path)
+    return tokens[0] if tokens else None
+
+
 def token_frequencies(posts: Iterable[Iterable[str]]) -> Counter[str]:
     """How many of ONE ARTIST's POSTS each working-name token appears in.