Implements production GPTQ quantization

[JyotinderSingh]

Presently the GPTQ algorithm simulates quantization and does not offer runtime benefits. This PR productionizes the code to implement runtime dequantization and resource savings

[codecov-commenter]

Codecov Report

❌ Patch coverage is 87.41722% with 19 lines in your changes missing coverage. Please review.
✅ Project coverage is 82.45%. Comparing base (b9ff57a) to head (7d114a6).

Files with missing lines	Patch %	Lines
keras/src/quantizers/gptq.py	88.70%	6 Missing and 1 partial ⚠️
keras/src/layers/core/einsum_dense.py	82.85%	4 Missing and 2 partials ⚠️
keras/src/layers/core/dense.py	87.50%	0 Missing and 3 partials ⚠️
keras/src/dtype_policies/dtype_policy.py	33.33%	1 Missing and 1 partial ⚠️
keras/src/layers/layer.py	80.00%	1 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##           master   #21610   +/-   ##
=======================================
  Coverage   82.45%   82.45%           
=======================================
  Files         572      572           
  Lines       57337    57431   +94     
  Branches     8970     8986   +16     
=======================================
+ Hits        47277    47357   +80     
- Misses       7761     7770    +9     
- Partials     2299     2304    +5     

Flag	Coverage Δ
keras	82.26% <86.75%> (+<0.01%)	⬆️
keras-jax	63.61% <86.09%> (+0.03%)	⬆️
keras-numpy	57.78% <16.55%> (-0.09%)	⬇️
keras-openvino	34.30% <14.56%> (-0.05%)	⬇️
keras-tensorflow	64.25% <86.09%> (+0.03%)	⬆️
keras-torch	63.82% <86.75%> (+0.03%)	⬆️

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.