add float8 and bfloat8 conversion functions

[chloeee99]

No description provided.

[tencent-adm]

Thank you for your submission, we really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.

---

chloeeyi seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[codecov-commenter]

Codecov Report

❌ Patch coverage is 0% with 30 lines in your changes missing coverage. Please review.
✅ Project coverage is 92.77%. Comparing base (6b4f6f9) to head (ed12b48).
⚠️ Report is 2 commits behind head on master.

Files with missing lines	Patch %	Lines
src/mat.cpp	0.00%	30 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #6495      +/-   ##
==========================================
- Coverage   93.27%   92.77%   -0.50%     
==========================================
  Files         845      807      -38     
  Lines      266119   255130   -10989     
==========================================
- Hits       248222   236702   -11520     
- Misses      17897    18428     +531     

☔ 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.

[Copilot]

Pull request overview

This PR adds float8 and bfloat8 conversion functions to support 8-bit floating-point formats. The implementation provides bidirectional conversions between float16 and two 8-bit formats: E4M3 (1-bit sign, 4-bit exponent, 3-bit mantissa) and E5M2 (1-bit sign, 5-bit exponent, 2-bit mantissa).

Key changes:

• Added float16 ↔ float8 E4M3 conversion functions with proper handling of special values
• Added float16 ↔ bfloat8 E5M2 inline conversion functions using direct bit truncation/extension
• Implemented proper handling for edge cases including zero, infinity, NaN, overflow, and underflow

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

File	Description
src/mat.h	Added function declarations and inline implementations for float8 and bfloat8 conversion functions
src/mat.cpp	Implemented float16_to_float8 and float8_to_float16 with comprehensive special value handling

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[nihui]

Thanks for your contribution !