use filename as document id when importing dataset

[dridk]

Hi,

I imported text data file into doccano for NER annotations. Filenames looks like "242424.txt".
When importing, used ids for document was an autoincrement number. It would be better to use the filename as id.
Without this feature, I cannot map my annotation with my previous files.

[github-actions]

Would you write your environment? Thank you!

[dridk]

red hat 9 / Last docker version