Apply suggestions from code review

Co-authored-by: Sam Abbott <contact@samabbott.co.uk>

Author: nikosbosse

R/metrics-nominal.R

@@ -73,8 +73,7 @@ assert_input_categorical <- function(
     summed_predictions <- round(rowSums(predicted, na.rm = TRUE), 10) # avoid numeric errors
   }
   # Allow for numeric errors
-  tolerance <- 1e-4
-  invalid_rows <- abs(summed_predictions - 1) > tolerance
+  invalid_rows <- abs(summed_predictions - 1) > 1e-4
   if (any(invalid_rows)) {
     #nolint start: keyword_quote_linter object_usage_linter
     row_indices <- which(invalid_rows)