Fix CI: bump mypy pin for numpy 2.5 stubs, sync confection pin

mypy 1.5.x crashes with an internal error on numpy>=2.3 type stubs,
which killed the mypy step on Python 3.12+. Bump to mypy 1.20.x and
fix the type errors the newer mypy reports.

Also sync the confection pin in requirements.txt with setup.cfg
(>=1.3.2), which was the single test failure on Python 3.10/3.11.

requirements.txt

@@ -28,11 +28,11 @@ pytest>=5.2.0,!=7.1.0
 pytest-timeout>=1.3.0,<2.0.0
 mock>=2.0.0,<3.0.0
 hypothesis>=3.27.0,<7.0.0
-mypy>=1.5.0,<1.6.0; platform_machine != "aarch64" and python_version >= "3.8"
+mypy>=1.20.2,<1.21.0; platform_machine != "aarch64" and python_version >= "3.8"
 types-mock>=0.1.1
 types-setuptools>=57.0.0
 types-requests
 types-setuptools>=57.0.0
 ruff>=0.9.0
 cython-lint>=0.15.0
-confection>=1.1.0,<2.0.0
+confection>=1.3.2,<2.0.0

spacy/cli/debug_data.py

@@ -579,7 +579,7 @@ def debug_data(
 
     if "morphologizer" in factory_names:
         msg.divider("Morphologizer (POS+Morph)")
-        label_list = [label for label in gold_train_data["morphs"]]
+        label_list = tuple(gold_train_data["morphs"])
         model_labels = _get_labels_from_model(nlp, "morphologizer")
         msg.info(f"{len(label_list)} label(s) in train data")
         labels = set(label_list)

spacy/cli/profile.py

@@ -66,7 +66,7 @@ def profile(model: str, inputs: Optional[Path] = None, n_texts: int = 10000) ->
 
         with msg.loading("Loading IMDB dataset via ml_datasets..."):
             imdb_train, _ = ml_datasets.imdb(train_limit=n_texts, dev_limit=0)
-            texts, _ = zip(*imdb_train)
+            texts = [text for text, _ in imdb_train]
         msg.info(f"Loaded IMDB dataset and using {n_texts} examples")
     with msg.loading(f"Loading pipeline '{model}'..."):
         nlp = load_model(model)

spacy/displacy/render.py

@@ -388,7 +388,7 @@ class DependencyRenderer:
             lang=self.lang,
         )
 
-    def render_word(self, text: str, tag: str, lemma: str, i: int) -> str:
+    def render_word(self, text: str, tag: str, lemma: Optional[str], i: int) -> str:
         """Render individual word.
 
         text (str): Word text.

spacy/lang/ja/__init__.py

@@ -61,7 +61,7 @@ class JapaneseTokenizer(DummyTokenizer):
             zip(*dtokens) if dtokens else [[]] * 7
         )
         sub_tokens_list = list(sub_tokens_list)
-        doc = Doc(self.vocab, words=words, spaces=spaces)
+        doc = Doc(self.vocab, words=list(words), spaces=spaces)
         next_pos = None  # for bi-gram rules
         for idx, (token, dtoken) in enumerate(zip(doc, dtokens)):
             token.tag_ = dtoken.tag

spacy/language.py

@@ -605,7 +605,7 @@ class Language:
                 existing_func = registry.factories.get(internal_name)
                 closure = existing_func.__closure__
                 wrapped = [c.cell_contents for c in closure][0] if closure else None
-                if util.is_same_func(wrapped, component_func):
+                if wrapped is not None and util.is_same_func(wrapped, component_func):
                     factory_func = existing_func  # noqa: F811
 
             cls.factory(

spacy/pipeline/spancat.py

@@ -250,7 +250,7 @@ class SpanCategorizer(TrainablePipe):
 
         DOCS: https://spacy.io/api/spancategorizer#init
         """
-        self.cfg = {
+        self.cfg: Dict[str, Any] = {
             "labels": [],
             "spans_key": spans_key,
             "threshold": threshold,

spacy/schemas.py

@@ -242,10 +242,10 @@ class TokenPatternNumber(BaseModel):
 
 
 class TokenPatternOperatorSimple(str, Enum):
-    plus: StrictStr = StrictStr("+")
-    star: StrictStr = StrictStr("*")
-    question: StrictStr = StrictStr("?")
-    exclamation: StrictStr = StrictStr("!")
+    plus = StrictStr("+")
+    star = StrictStr("*")
+    question = StrictStr("?")
+    exclamation = StrictStr("!")
 
 
 TokenPatternOperatorMinMax = constr(pattern=r"^(\{\d+\}|\{\d+,\d*\}|\{\d*,\d+\})$")