fix: remove uint/uint64 from integer arithmetic to avoid overflow concerns

The gosec linter (G115) flagged potential integer overflow when converting
uint and uint64 to int64. To address this, removed these types from
integer arithmetic support, aligning with the divided_by filter pattern.

These types will now fall back to float arithmetic, which is safe and
matches the existing pattern in the codebase.

Author: claude

filters/standard_filters.go

@@ -28,9 +28,11 @@ type FilterDictionary interface {
 // Helper functions for type-aware arithmetic operations
 
 // isIntegerType checks if a value is an integer type
+// Note: uint and uint64 are not included to avoid potential overflow issues
+// when converting to int64. These types will fall back to float arithmetic.
 func isIntegerType(v any) bool {
 	switch v.(type) {
-	case int, int8, int16, int32, int64, uint, uint8, uint16, uint32, uint64:
+	case int, int8, int16, int32, int64, uint8, uint16, uint32:
 		return true
 	default:
 		return false
@@ -50,16 +52,12 @@ func toInt64(v any) int64 {
 		return int64(val)
 	case int64:
 		return val
-	case uint:
-		return int64(val)
 	case uint8:
 		return int64(val)
 	case uint16:
 		return int64(val)
 	case uint32:
 		return int64(val)
-	case uint64:
-		return int64(val)
 	default:
 		return 0
 	}