False positive NDArray in Jupyter notebook

[eismont21]

It was fixed for the .py files #565, but it is stilling failing in Jupyter notebooks.

[epage]

Currently, none of our language-specific terms recognize Jupyter, see https://github.com/crate-ci/typos/blob/master/crates/typos-cli/src/file_type_specifics.rs

iirc Jupyter allows multiple languages in a notebook, making them ore like Markdown, rather than having a way to tell by extension. Is that right?

What we'd need to address include

• Should naively handle jupyter like we do for markdown (see also Ratatui #1055)?
• If we handle it, should we assume its mostly Python people and just make it a synonym for Python?
• If we want to handle these "cross-language" files, how far do we take that? For example, you could have comments in Python code referring to C code that you are using bindings for.