refactor(setup): move setup_* helpers to a separate module

[cxw620]

Part of #62

---

Breaking changes:

• No longer export TlsCryptoInfo (formerly CryptoInfo)
• config_ktls_* now requires the socket implements AsFd.

[codecov]

Codecov Report

❌ Patch coverage is 0% with 153 lines in your changes missing coverage. Please review.
✅ Project coverage is 28.76%. Comparing base (571d49a) to head (4ec55cb).
⚠️ Report is 20 commits behind head on main.

Files with missing lines	Patch %	Lines
ktls/src/setup/tls.rs	0.00%	103 Missing ⚠️
ktls/src/lib.rs	0.00%	38 Missing ⚠️
ktls/src/setup/ulp.rs	0.00%	11 Missing ⚠️
ktls/src/ffi.rs	0.00%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #66      +/-   ##
==========================================
+ Coverage   24.74%   28.76%   +4.01%     
==========================================
  Files           6        8       +2     
  Lines        1265      970     -295     
==========================================
- Hits          313      279      -34     
+ Misses        952      691     -261     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

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

[djc]

The commit history here doesn't currently seem to make much sense. The first commit seems to make stylistic changes and adds a bunch of new code (with bullet points in the commit message, suggesting that there are things that could be separate commits), and the second commit only removing a bunch of code (which apparently has become useless -- if that was caused by the first commit, it should be in the same commit).

[cxw620]

While libc is upgraded to 0.2.175, ffi::TLS_1_2_VERSION_NUMBER and ffi::TLS_1_3_VERSION_NUMBER can also be replaced with libc::TLS_1_2_VERSION and libc::TLS_1_3_VERSION, then this crate will no longer need the ktls-sys dependency. Given that it's unrelated to this PR, it will be completed in the next PR.

[djc]

Okay, this still looks like very far from what I consider to be commits that are easy to review. I'm going to give up on reviewing this PR.