add mixed precision training/matmul32 precision

[jveitchmichaelis]

Description

Adds config options to enabled mixed precision training via Lightning, and allow setting matmul32 precision on CUDA systems. This suppresses a warning from torch when CUDA cores are available, and also provides a decent speedup.

The defaults are auto equivalents to avoid any issues with existing models, but I've seen pretty good speedups using 16-mixed/high

AI-Assisted Development

• I used AI tools (e.g., GitHub Copilot, ChatGPT, etc.) in developing this PR
• I understand all the code I'm submitting
• I have reviewed and validated all AI-generated code

AI tools used (if applicable):
Checks to make sure configs are consistent.

[codecov]

Codecov Report

❌ Patch coverage is 66.66667% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 87.35%. Comparing base (884502e) to head (6c14902).
⚠️ Report is 5 commits behind head on main.

Files with missing lines	Patch %	Lines
src/deepforest/main.py	50.00%	2 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1356   +/-   ##
=======================================
  Coverage   87.35%   87.35%           
=======================================
  Files          24       24           
  Lines        2981     3014   +33     
=======================================
+ Hits         2604     2633   +29     
- Misses        377      381    +4     

Flag	Coverage Δ
unittests	87.35% <66.66%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ 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.