Generate realistic seed data that covers edge cases

You are a QA-minded backend engineer. Generate a realistic seed data script for testing a database table or set of tables.

Database engine: {{DB_ENGINE}}
Schema (DDL or description):
```
{{SCHEMA}}
```
Application context (what does this data represent and what does the app do with it):
{{APP_CONTEXT}}
Number of seed rows per main table: {{ROW_COUNT}}

Complete the following:
1. Identify the columns most likely to cause bugs: nullable fields, enums, date ranges, foreign keys with optional references, numeric fields near zero or at type limits, and free-text fields.
2. Ensure the seed data deliberately includes at least one row for each of these edge cases: null optional fields, maximum-length strings, dates at boundaries (epoch, far future, Feb 29), zero and negative numbers where the type allows, and duplicate values where uniqueness is not enforced.
3. Include a realistic distribution of 'normal' rows alongside the edge-case rows so the dataset feels plausible.
4. Output the seed data as executable SQL INSERT statements (or a {{DB_ENGINE}}-compatible seed file format).
5. Add a comment above each edge-case row explaining what it is testing.
6. List any constraints you could not satisfy due to schema rules, and explain why.