[fix](doc) trim-in.md en: trim('ababccaab','ab') now returns 'cca', not 'ababccaab'

[boluor]

Summary

Doc page (4.x): `scalar-functions/string-functions/trim-in.md` (EN).

The "Comparison with TRIM" example claimed that:

```sql
SELECT trim_in('ababccaab', 'ab'), trim('ababccaab', 'ab');
```

returns:

trim_in	trim
cc	ababccaab

On Apache Doris 4.1.1 the second column is `cca`, not the input unchanged. `trim(, )` peels the literal `` from both ends and repeats until neither end matches, so `'ababccaab'` peels `'ab'` twice from the front and once from the back, leaving `'cca'`.

The side-by-side example's pedagogical point still works — `trim_in` treats the 2nd arg as a character set (peels any `a` or `b`) while `trim` treats it as a literal substring (peels exact `ab`) — just with different output. Update the result row to match cluster reality and tighten the intro prose to call out the char-set-vs-substring contrast more clearly. See PR #3810 for the companion fix on trim.md's own page.

Verification

```
mysql> SELECT trim_in('ababccaab', 'ab'), trim('ababccaab', 'ab');
+----------------------------+-------------------------+
| trim_in('ababccaab', 'ab') | trim('ababccaab', 'ab') |
+----------------------------+-------------------------+
| cc | cca |
+----------------------------+-------------------------+
```

Test plan

• Run on a 4.1.1 cluster — matches the new result row.
• Divider widths updated to fit the shorter values.
• No other change.

🤖 Generated with Claude Code