Add g:ncm2_jedi#settings

[theFool32]

Some packages like numpy, torch, tensorflow are so large and it takes a long time to complete them.
Anyway to cache it ahead?
Thanks.

[antonzub1]

@roxma, first of all - thank you very much for doing great job and delivering such a good product. It works really well, except speed of completion on large packages, as @theFool32 mentioned. I am wondering - is it possible to utilize Jedi cache somehow to make completions more fast?

[roxma]

Thanks for the info @zabulazza

I will get some free time and test it.

[roxma]

@zabulazza

filesystem-cache is enabled by default.

I'm not sure what I can do for this.

You could try this fork for now: #6

I'm going to add g:ncm2_jedi#settings for customized settings for https://jedi.readthedocs.io/en/latest/docs/settings.html and then close this issue.