fix(types): annotate rank_rules_for_draft to satisfy pyright

src/gradata/hooks/jit_inject.py

@@ -25,10 +25,14 @@
 import re
 import time
 from pathlib import Path
+from typing import TYPE_CHECKING
 
 from gradata.hooks._base import extract_message, resolve_brain_dir, run_hook
 from gradata.hooks._profiles import Profile
 
+if TYPE_CHECKING:
+    from gradata._types import Lesson
+
 try:
     from gradata.enhancements.self_improvement import parse_lessons
 except ImportError:
@@ -96,13 +100,13 @@ def _float_env(name: str, default: float) -> float:
 
 
 def rank_rules_for_draft(
-    lessons: list,
+    lessons: list[Lesson],
     draft_text: str,
     *,
     k: int = DEFAULT_MAX_RULES,
     min_confidence: float = DEFAULT_MIN_CONFIDENCE,
     min_similarity: float = DEFAULT_MIN_SIMILARITY,
-) -> list[tuple[object, float]]:
+) -> list[tuple[Lesson, float]]:
     """Score each lesson against draft_text and return top-k above threshold.
 
     Returns a list of (lesson, similarity) tuples, highest first. A rule
@@ -117,7 +121,7 @@ def rank_rules_for_draft(
     if not draft_tokens:
         return []
 
-    scored: list[tuple[object, float]] = []
+    scored: list[tuple[Lesson, float]] = []
     for lesson in lessons:
         conf = getattr(lesson, "confidence", 0.0)
         if conf < min_confidence:

src/gradata/rules/rule_context.py

@@ -64,7 +64,7 @@ def _rule_matches_domain(rule: GraduatedRule, domain_norm: str) -> bool:
     if str(scope.get("domain", "")).strip().lower() == domain_norm:
         return True
     applies = str(scope.get("applies_to", "")).strip().lower()
-    return applies == domain_norm or (applies and applies.startswith(f"{domain_norm}:"))
+    return applies == domain_norm or applies.startswith(f"{domain_norm}:")
 
 
 class RuleContext: