feat(quando-r1o): consolidate benchmarks and achieve 99.5% test coverage

Reorganized test suite with consolidated benchmarks and comprehensive
edge case coverage, exceeding 95% minimum coverage requirement.

Changes:
- Created benchmark_test.go with all 52 benchmarks organized by category
  (arithmetic, clock, date, diff, format, inspection, parse, snap, unit)
- Removed benchmarks from 9 individual test files for better organization
- Added 6 edge case tests to improve coverage:
  * Format() unknown type fallback
  * formatLong() empty lang default
  * Format.String() unknown value
  * isYearPrefix() edge cases
  * StartOf()/EndOf() unsupported units
- Added 3 DST transition tests:
  * Spring forward (23-hour day)
  * Fall back (25-hour day)
  * Multiple timezone preservation

Results:
- Test coverage: 97.7% → 99.5% (exceeds 95% target)
- All 52 benchmarks consolidated and verified
- All benchmarks meet performance targets
- No test regressions

Files modified:
- Created: benchmark_test.go (580 lines)
- Modified: 9 test files (removed benchmarks, added tests)

unit_test.go

@@ -136,12 +136,3 @@ func TestUnitComparability(t *testing.T) {
 	}
 }
 
-// BenchmarkUnitString benchmarks the String() method
-func BenchmarkUnitString(b *testing.B) {
-	units := []Unit{Seconds, Minutes, Hours, Days, Weeks, Months, Quarters, Years}
-	b.ResetTimer()
-	for i := 0; i < b.N; i++ {
-		u := units[i%len(units)]
-		_ = u.String()
-	}
-}