Plan the cleanup of a messy column

Free-text data that half-works and needs to become reliable.

Below is a sample of a messy column. Plan its cleanup.

1. Classify what you see into distinct patterns, with a count or share for each, and give an example of each pattern.
2. Name the ambiguous cases where you genuinely cannot tell the intended value, and say what a human needs to decide.
3. Propose a normalisation order, cheapest and safest first: trim, case-fold, strip punctuation, collapse whitespace, then the harder mappings.
4. Give the SQL or regex for each step, and for each, an example of a value it would wrongly change.
5. Recommend whether to clean on read, clean on write, or add a cleaned column alongside the original. Give one answer.

Never silently discard values you cannot parse. Route them to a quarantine bucket and tell me how to count it.

Sample values:
What the column is supposed to hold:

Schema and data quality. Designing tables, migrating them without breaking live code, and cleaning what you have been handed.