feat: split long input into chunks for translation

[min0625]

Summary

• Long input (over 2000 runes) is split at paragraph boundaries (falling back to line breaks, then whitespace) and translated chunk by chunk, so no single request can hit the model's output-token limit and come back truncated
• Each chunk's original separator (paragraph gap, line break, single space, or nothing) is reprinted verbatim between chunk outputs, so internal whitespace survives exactly regardless of how the document was split
• Rotation-mode language detection samples a single representative chunk instead of the whole document, so detection cost doesn't grow with document length; a neutral sample only short-circuits output for single-chunk documents
• Marks "Batch translation mode" as done in the README roadmap

Test plan

• make check (lint + test) passes via pre-commit hook
• New unit tests in cmd/mint/chunk_test.go and cmd/mint/main_test.go cover chunk splitting/packing and end-to-end chunked translation
• Manual test case added in docs/manual-testing.md (fix: normalize semver pre-release version to PEP 440 in build_wheels.py #15) for chunked translation over a multi-paragraph document

🤖 Generated with Claude Code

[qodo-code-review]

Qodo reviews are paused for this user.

Troubleshooting steps vary by plan Learn more →

On a Teams plan?
Reviews resume once this user has a paid seat and their Git account is linked in Qodo.
Link Git account →

Using GitHub Enterprise Server, GitLab Self-Managed, or Bitbucket Data Center?
These require an Enterprise plan - Contact us
Contact us →

[codecov]

Codecov Report

❌ Patch coverage is 90.44944% with 17 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
cmd/mint/main.go	85.52%	6 Missing and 5 partials ⚠️
cmd/mint/chunk.go	93.93%	3 Missing and 3 partials ⚠️

📢 Thoughts on this report? Let us know!