Many Chinese and Japanese HTML and Markdown documents today abuse a ASCII space to represent a space between han and alphanumerics.

https://github.com/huacnlee/autocorrect?tab=readme-ov-file#configuration
https://prettier.io/blog/2023/07/05/3.0.0.html#stop-inserting-spaces-between-chinese-or-japanese-and-western-characters
https://github.com/lint-md/lint-md?tab=readme-ov-file#%E8%A7%84%E5%88%99%E6%A6%82%E8%BF%B0
https://github.com/textlint-ja/textlint-rule-preset-ja-spacing/tree/master/packages/textlint-rule-ja-space-between-half-and-full-width#options

It makes it hard to search a text from a document. An ASCII space is not strictly the same width as 0.25ic, which is stipulated in jlreq or clreq. (only clreq allows an ASCII space) Therefore, documents abusing an ASCII space will not migrate to text-autospace because the width by text-autospace it too narrow.

Such ASCII spaces have polluted outputs by LLM. LLM tends to output text including such spaces. Translations produced by AI in Microsoft contain a lot of such spaces (e.g. in Visual Studio (Code)).

text-autospace should allow adjustment of the width.