mlx - nn and core updates post merge

[acsweet]

This addresses #19571

Patched and implemented missing methods in:

• nn.py (a lot has been added since the mlx branch started!)
• core.py
• rnn.py (very simple scan update)

Started to adjust tests for various layers that mlx doesn't support:

• flash_attention
• dtypes float64, complex128

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 0% with 225 lines in your changes missing coverage. Please review.

Please upload report for BASE (mlx@168f13d). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
keras/src/backend/mlx/core.py	0.00%	129 Missing ⚠️
keras/src/backend/mlx/nn.py	0.00%	93 Missing ⚠️
keras/src/backend/mlx/__init__.py	0.00%	3 Missing ⚠️

Additional details and impacted files

@@          Coverage Diff           @@
##             mlx   #20832   +/-   ##
======================================
  Coverage       ?   72.72%           
======================================
  Files          ?      572           
  Lines          ?    55023           
  Branches       ?     8489           
======================================
  Hits           ?    40016           
  Misses         ?    13069           
  Partials       ?     1938           

Flag	Coverage Δ
keras	72.64% <0.00%> (?)
keras-jax	61.09% <0.00%> (?)
keras-numpy	56.08% <0.00%> (?)
keras-openvino	28.36% <0.00%> (?)
keras-tensorflow	61.60% <0.00%> (?)

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.

[fchollet]

LGTM, thank you!