把每篇文章的标签列表展开：每个 (article_id, 标签) 一行。
- 空列表和 null 列表都不产生任何行。
- 同一篇文章里重复出现的标签照样输出多行（不去重），大小写也保持原样。
- 列表里的元素本身不会是 null。
输入
articles(article_id, title, author, tags, published_at, word_count)：tags是字符串列表，可能为空列表或 null
输出
列为 article_id, tag，行顺序不限。
展开标签
输入表 （case base，共 2 个 case：base、edge）
articles blog/articles
| # | article_idint64 | titlestring | authorstruct<name: string, country: string> | tagslist<string> | published_atdatetime | word_countint64 |
|---|---|---|---|---|---|---|
| 1 | 1 | A gentle intro to DuckDB | {"name": "Hans Muller", "country": "DE"} | ["etl", "Python", "rust"] | 2026-02-07 18:20:00 | 353 |
| 2 | 2 | Testing Python | {"name": "Bob Smith", "country": "US"} | ["SQL"] | 2026-09-02 17:13:00 | 3726 |
| 3 | 3 | 10 tips for dataframes | {"name": "Bob Smith", "country": "US"} | ["pandas"] | 2026-07-23 20:39:00 | 362 |
| 4 | 4 | Debugging DuckDB | {"name": "Amy Chen", "country": "US"} | [] | 2026-01-20 22:03:00 | 3845 |
| 5 | 5 | Debugging window functions | {"name": "Ana Souza", "country": "BR"} | [] | 2026-08-20 10:17:00 | 3899 |
| 6 | 6 | Getting started with Polars | {"name": "王芳", "country": "CN"} | ["polars"] | 2026-05-17 19:28:00 | 2086 |
| 7 | 7 | Testing dataframes | {"name": "Ana Souza", "country": "BR"} | ["rust", "ml", "go"] | 2026-08-01 17:58:00 | 1251 |
| 8 | 8 | Benchmarking pandas | {"name": "王芳", "country": "CN"} | ["Python", "pandas", "ml", "go"] | 2026-02-27 20:10:00 | 2984 |
| 9 | 9 | Benchmarking Python | {"name": "Hans Muller", "country": "DE"} | ["etl"] | 2026-09-09 09:30:00 | 1709 |
| 10 | 10 | 10 tips for dataframes | {"name": "Amy Chen", "country": "US"} | null | 2026-08-11 21:07:00 | 2000 |
| 11 | 11 | Debugging pandas | {"name": "Amy Chen", "country": "US"} | ["python", "viz"] | 2026-06-23 17:24:00 | 3908 |
| 12 | 12 | Getting started with SQL | {"name": "Ana Souza", "country": "BR"} | ["sql"] | 2026-03-01 18:14:00 | 2493 |
| 13 | 13 | Common pitfalls in SQL | {"name": "Li Wei", "country": "CN"} | ["python", "pandas", "python"] | 2026-03-25 19:46:00 | 473 |
| 14 | 14 | A gentle intro to ETL | {"name": "王芳", "country": "CN"} | ["viz", "SQL", "etl", "go"] | 2026-07-21 18:59:00 | 3850 |
| 15 | 15 | Common pitfalls in Python | {"name": "Amy Chen", "country": "US"} | ["SQL", "polars"] | 2026-06-08 15:27:00 | 1357 |
| 16 | 16 | Getting started with ETL | {"name": "Ana Souza", "country": "BR"} | ["rust", "ml", "python"] | 2026-07-16 18:30:00 | 2713 |
| 17 | 17 | A gentle intro to pandas | {"name": "Kenji Sato", "country": "JP"} | ["Python", "ml"] | 2026-05-27 20:35:00 | 2305 |
| 18 | 18 | Getting started with pandas | {"name": "Li Wei", "country": "CN"} | null | 2026-06-17 11:01:00 | 1790 |
| 19 | 19 | A gentle intro to ETL | {"name": "Kenji Sato", "country": "JP"} | ["duckdb", "SQL", "viz"] | 2026-02-27 21:10:00 | 3630 |
| 20 | 20 | Benchmarking DuckDB | {"name": "Li Wei", "country": "CN"} | ["Python", "duckdb"] | 2026-05-07 13:35:00 | 1067 |
显示前 20 行，共 40 行
期望输出 （case base，行顺序不限）
| # | article_idInt64 | tagString |
|---|---|---|
| 1 | 1 | etl |
| 2 | 1 | Python |
| 3 | 1 | rust |
| 4 | 2 | SQL |
| 5 | 3 | pandas |
| 6 | 6 | polars |
| 7 | 7 | rust |
| 8 | 7 | ml |
| 9 | 7 | go |
| 10 | 8 | Python |
| 11 | 8 | pandas |
| 12 | 8 | ml |
| 13 | 8 | go |
| 14 | 9 | etl |
| 15 | 11 | python |
| 16 | 11 | viz |
| 17 | 12 | sql |
| 18 | 13 | python |
| 19 | 13 | pandas |
| 20 | 13 | python |
| 21 | 14 | viz |
| 22 | 14 | SQL |
| 23 | 14 | etl |
| 24 | 14 | go |
| 25 | 15 | SQL |
| 26 | 15 | polars |
| 27 | 16 | rust |
| 28 | 16 | ml |
| 29 | 16 | python |
| 30 | 17 | Python |
| 31 | 17 | ml |
| 32 | 19 | duckdb |
| 33 | 19 | SQL |
| 34 | 19 | viz |
| 35 | 20 | Python |
| 36 | 20 | duckdb |
| 37 | 21 | ml |
| 38 | 21 | Python |
| 39 | 22 | polars |
| 40 | 22 | rust |
| 41 | 22 | etl |
| 42 | 23 | sql |
| 43 | 23 | duckdb |
| 44 | 23 | sql |
| 45 | 23 | SQL |
| 46 | 24 | duckdb |
| 47 | 25 | python |
| 48 | 25 | viz |
| 49 | 26 | SQL |
| 50 | 26 | polars |
显示前 50 行，共 74 行