add ops.layer_normalization

[pass-lin]

from #21313 and #21323
To improve the performance of the torch backend, we add a new op and use the cudnn operator of the torch in this op

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 87.93103% with 7 lines in your changes missing coverage. Please review.

Project coverage is 82.59%. Comparing base (240d4f4) to head (48fb1c6).

Files with missing lines	Patch %	Lines
keras/src/ops/nn.py	90.56%	3 Missing and 2 partials ⚠️
keras/api/_tf_keras/keras/ops/__init__.py	0.00%	1 Missing ⚠️
keras/api/_tf_keras/keras/ops/nn/__init__.py	0.00%	1 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##           master   #21327   +/-   ##
=======================================
  Coverage   82.58%   82.59%           
=======================================
  Files         565      565           
  Lines       54819    54850   +31     
  Branches     8516     8520    +4     
=======================================
+ Hits        45273    45303   +30     
+ Misses       7454     7452    -2     
- Partials     2092     2095    +3     

Flag	Coverage Δ
keras	82.40% <86.20%> (+<0.01%)	⬆️
keras-jax	63.53% <81.03%> (+<0.01%)	⬆️
keras-numpy	58.69% <74.13%> (+<0.01%)	⬆️
keras-openvino	33.11% <58.62%> (-0.01%)	⬇️
keras-tensorflow	63.94% <81.03%> (+<0.01%)	⬆️
keras-torch	63.59% <63.79%> (-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.

[fchollet]

LGTM, thanks for the refactor.