feat(async): Migrate Together.ai provider to async

[besaleli]

Description

NOTE: Untested until I get my hands on a Together.ai API key 🥀 they rate limited me

PR Type: 🆕 New Feature

Relevant issues

Closes #217

Checklist

• I have added unit tests that prove my fix/feature works
• New and existing tests pass locally
• Documentation was updated where necessary
• I have read and followed the contribution guidelines

[codecov]

Codecov Report

❌ Patch coverage is 20.00000% with 28 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
src/any_llm/providers/together/together.py	21.73%	18 Missing ⚠️
src/any_llm/providers/together/utils.py	16.66%	10 Missing ⚠️

Files with missing lines	Coverage Δ
src/any_llm/providers/together/utils.py	17.07% <16.66%> (-2.93%)	⬇️
src/any_llm/providers/together/together.py	40.98% <21.73%> (-3.25%)	⬇️

... and 26 files with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[njbrake]

This is 10000% scope creep but if you feel like migrating this to the utils file, that would probably be a good improvement

[anivar]

I actually implemented this suggestion while working on the same issue. The duplicated code (lines 215-249) can be extracted to a to_chat_completion utility function in utils.py. This would make both methods cleaner and easier to maintain. Would you like me to push a commit with this refactoring to your branch?

[anivar]

I noticed the response conversion logic is duplicated between completion and acompletion methods. Consider extracting it to a utility function to reduce duplication. Happy to help if needed!

[besaleli]

Hey @anivar, thanks for the comment. Feel free to fork off this branch (217-migrate-together-provider-to-async) and submit a PR to merge onto this branch if you want to take care of the utils refactor. We will keep this PR as the source of truth for the final enhancement integration.

[anivar]

Thanks @besaleli! Since I can't push directly to your branch, I've created PR #272 with the utility function refactoring as suggested by @njbrake.

The refactor extracts the duplicated response conversion logic and adds the missing reasoning_effort handling for consistency. Feel free to merge it into your branch if it looks good!

[njbrake]

LGTM 🚀