Refactor reparameterizable layers and add GhostNetV3

[james77777778]

related to #2

[codecov]

Codecov Report

Attention: Patch coverage is 98.92473% with 3 lines in your changes are missing coverage. Please review.

Project coverage is 94.78%. Comparing base (e2d31cc) to head (79c44ff).

Files	Patch %	Lines
kimm/_src/utils/timm_utils.py	0.00%	2 Missing ⚠️
kimm/_src/layers/reparameterizable_conv2d.py	98.75%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #47      +/-   ##
==========================================
+ Coverage   94.77%   94.78%   +0.01%     
==========================================
  Files          39       39              
  Lines        4266     4295      +29     
==========================================
+ Hits         4043     4071      +28     
- Misses        223      224       +1     

Flag	Coverage Δ
kimm	94.78% <98.92%> (+0.01%)	⬆️
kimm-jax	94.13% <98.92%> (+0.01%)	⬆️
kimm-numpy	63.98% <52.68%> (-2.39%)	⬇️
kimm-tensorflow	94.41% <98.20%> (-0.04%)	⬇️
kimm-torch	94.13% <98.92%> (+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.