CSV cleanup and duplicate detection — demonstration

Synthetic-data demonstration of a reproducible cleanup workflow. Product identifiers and names are trimmed, repeated whitespace is normalized, and status values use consistent lowercase text. Duplicate A02 records are flagged for review rather than silently deleted.

The example preserves all four source rows and identifies the two duplicate rows. Deliverables for a scoped client project can include the cleaned file, duplicate review list, validation checks, and change log. This is an Arratz demonstration, not a client commission.