Investigating the Effectiveness of Equating Designs for Constructed‐Response Tests in Large‐Scale Assessments