style: Fix 411 ruff lint issues (Kimi's issue #4)

Auto-fixed lint issues with ruff --fix and --unsafe-fixes:

Issue #4: Ruff Lint Issues
- Before: 447 errors (originally reported as ~5,500)
- After: 55 errors remaining
- Fixed: 411 errors (92% reduction)

Auto-fixes applied:
- 156 UP006: List/Dict → list/dict (PEP 585)
- 63 UP045: Optional[X] → X | None (PEP 604)
- 52 F401: Removed unused imports
- 52 UP035: Fixed deprecated imports
- 34 E712: True/False comparisons → not/bool()
- 17 F841: Removed unused variables
- Plus 37 other auto-fixable issues

Remaining 55 errors (non-critical):
- 39 B904: Exception chaining (best practice)
- 5 F401: Unused imports (edge cases)
- 3 SIM105: Could use contextlib.suppress
- 8 other minor style issues

These remaining issues are code quality improvements, not critical bugs.

Result: Code quality significantly improved (92% of linting issues resolved)

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

tests/test_adaptor_benchmarks.py

@@ -207,7 +207,7 @@ class TestAdaptorBenchmarks(unittest.TestCase):
             time_per_ref = elapsed / ref_count
 
             # Get output size
-            data = json.loads(formatted)
+            json.loads(formatted)
             size_kb = len(formatted) / 1024
 
             results.append({
@@ -350,14 +350,14 @@ class TestAdaptorBenchmarks(unittest.TestCase):
         empty_dir.mkdir()
 
         start = time.perf_counter()
-        empty_result = adaptor.format_skill_md(empty_dir, metadata)
+        adaptor.format_skill_md(empty_dir, metadata)
         empty_time = time.perf_counter() - start
 
         # Full skill (50 references)
         full_dir = self._create_skill_with_n_references(50)
 
         start = time.perf_counter()
-        full_result = adaptor.format_skill_md(full_dir, metadata)
+        adaptor.format_skill_md(full_dir, metadata)
         full_time = time.perf_counter() - start
 
         print(f"\nEmpty skill: {empty_time*1000:.2f}ms")