can't encode non-GBK Character into GB18030

**Describe the bug**
can't encode non-GBK Character (like most extended Latin letters and Emoji) into GB18030

**To Reproduce**
1. Choose "Encode text-Simplified Chinese GB18030";
2. Type non-GBK Character (like "ä" or "😀") in input;
3. Empty bytes would be returned.

**Expected behaviour**
non-GBK Character should be encoded into corresponding GB18030 bytes (4 bytes per character).

Reproduce the issue with the “Encode text-Simplified Chinese GB18030” operation using “ä” and “😀”, then inspect the operation entry point and its encoding dependencies. Done means non-GBK characters produce their corresponding GB18030 bytes instead of empty output, with regression coverage if the project’s existing tests identify a suitable location.

Written by the indexing model from the issue text.