feat: support MinerU official cloud API for PDF parsing

[wyuc]

Currently the MinerU provider only supports self-hosted instances (POST {baseUrl}/file_parse), which requires a GPU server.

MinerU now has an official cloud API at https://mineru.net/apiManage/docs with free tier (2000 pages/day). It uses a different async task model (POST /api/v4/extract/task → poll → download), so a new mineru-cloud provider is needed alongside the existing self-hosted one.

This would make MinerU's advanced PDF parsing (tables, formulas, layout analysis) accessible without any infrastructure setup.

[soni0021]

Hi @wyuc ,

This project looks really interesting. I have hands-on experience using MinerU for building complex data extraction and enhancement pipelines.

I’d love to work on this issue
could you please assign it to me?

[technoadnan]

hi @wyuc I address the fix of this issue yesterday #191