不能处理英文'

```python
from LAC import LAC
lac = LAC(mode="seg")
lac.run("I don't like it")
```
输出为
```
['I', ' ', 'don', "'", 't', ' ', 'like', ' ', 'it']
```

Start by reproducing the Python example through lac.run("I don't like it") and inspect the segmentation path responsible for English text. The issue shows that the apostrophe is split into a separate token; done means the intended English contraction behavior is handled consistently and the original example no longer produces the reported output.

Written by the indexing model from the issue text.