[html5lib] Annotate more (#14994)

* Import `webencodings` to use `Encoding` class instead of `Protocol`
* Make `_inputstream.BufferedStream`, `filters.base.Filter`, `_utils.MethodDispatcher` generic
* Fix incorrect types in `_ihatexml.pyi`
* Add types for some attributes/methods

stubs/bleach/bleach/html5lib_shim.pyi

@@ -50,7 +50,7 @@ class InputStreamWithMemory:
 
 class BleachHTMLTokenizer(HTMLTokenizer):
     consume_entities: bool
-    stream: InputStreamWithMemory
+    stream: InputStreamWithMemory  # type: ignore[assignment]
     emitted_last_token: dict[str, Any] | None
     def __init__(self, consume_entities: bool = False, **kwargs: Any) -> None: ...
 

stubs/bleach/bleach/linkifier.pyi

@@ -36,7 +36,7 @@ class Linker:
 # or `html5lib` token might be reused
 _Token: TypeAlias = dict[str, Any]
 
-class LinkifyFilter(Filter):
+class LinkifyFilter(Filter[_Token]):
     callbacks: Iterable[_Callback]
     skip_tags: Container[str]
     parse_email: bool

stubs/bleach/bleach/sanitizer.pyi

@@ -40,7 +40,7 @@ class Cleaner:
     protocols: Iterable[str]
     strip: bool
     strip_comments: bool
-    filters: Iterable[Filter]
+    filters: Iterable[_FilterConstructor]
     css_sanitizer: CSSSanitizer | None
     parser: BleachHTMLParser
     walker: TreeWalker
@@ -85,7 +85,7 @@ class BleachSanitizerFilter(SanitizerFilter):
     def sanitize_stream(self, token_iterator: Iterable[_Token]) -> Iterator[_Token]: ...
     def merge_characters(self, token_iterator: Iterable[_Token]) -> Iterator[_Token]: ...
     def __iter__(self) -> Iterator[_Token]: ...
-    def sanitize_token(self, token: _Token) -> _Token | list[_Token] | None: ...
+    def sanitize_token(self, token: _Token) -> _Token | list[_Token] | None: ...  # type: ignore[override]
     def sanitize_characters(self, token: _Token) -> _Token | list[_Token]: ...
     def sanitize_uri_value(self, value: str, allowed_protocols: Container[str]) -> str | None: ...
     def allow_token(self, token: _Token) -> _Token: ...