Repair scalar and array result normalization in three vendored Evaluate classification metrics.

Repair scalar and array result normalization in three vendored Evaluate classification metrics.: a task in MiMo-V2.6-RL-harbor-terminal: MiMo-V2.6-RL Terminal (Harbor) (Harbor dataset). The workspace contains a frozen slice of a Hugging Face Evaluate repository under /app/vendor/project . Its…