feat: Add Grain dataset support for keras.layers.TextVectorization.adapt()

[dataCenter430]

Summary

Adds support for Grain datasets in keras.layers.TextVectorization.adapt(), so users can adapt the layer on grain.MapDataset, grain.IterDataset, or grain.DataLoader without converting to tf.data.Dataset.

Closes #22378.

Changes

• adapt()
  • When data is a Grain dataset (grain.MapDataset, grain.IterDataset, or grain.DataLoader), it is converted to a tf.data.Dataset via GrainDatasetAdapter.get_tf_dataset() and then iterated as in the existing tf.data path.
  • steps is supported for Grain datasets (same semantics as for tf.data).
• Batch handling
  • Introduced _extract_adapt_batch(batch) so that when a batch is (x,), (x, y), or (x, y, sample_weight), only the text input x is passed to update_state().
  • This helper is used for both tf.data and Grain branches so adapt() works with training-style batches.
• Docs
  • Docstring of adapt() updated to mention Grain datasets and that batches may be a single tensor or a tuple (only text is used).
• Tests
  • test_adapt_with_grain_dataset: adapt on a Grain MapDataset and assert vocabulary.
  • test_adapt_with_grain_dataset_and_steps: adapt with steps=2 and assert only first batches are used.
  • Both tests use pytest.importorskip("grain") so they are skipped when Grain is not installed.

[gemini-code-assist]

Warning

Gemini encountered an error creating the summary. You can try again by commenting /gemini summary.

[codecov-commenter]

Codecov Report

❌ Patch coverage is 30.76923% with 9 lines in your changes missing coverage. Please review.
✅ Project coverage is 77.13%. Comparing base (d9966a5) to head (9492201).
⚠️ Report is 24 commits behind head on master.

Files with missing lines	Patch %	Lines
...ras/src/layers/preprocessing/text_vectorization.py	30.76%	7 Missing and 2 partials ⚠️

❗ There is a different number of reports uploaded between BASE (d9966a5) and HEAD (9492201). Click for more details.

HEAD has 2 uploads less than BASE

Flag	BASE (d9966a5)	HEAD (9492201)
keras	5	4
keras-tensorflow	1	0

Additional details and impacted files

@@            Coverage Diff             @@
##           master   #22402      +/-   ##
==========================================
- Coverage   82.96%   77.13%   -5.83%     
==========================================
  Files         596      596              
  Lines       66269    66696     +427     
  Branches    10321    10386      +65     
==========================================
- Hits        54982    51449    -3533     
- Misses       8667    12575    +3908     
- Partials     2620     2672      +52     

Flag	Coverage Δ
keras	77.01% <30.76%> (-5.79%)	⬇️
keras-jax	60.55% <30.76%> (-0.29%)	⬇️
keras-numpy	54.78% <30.76%> (-0.26%)	⬇️
keras-openvino	49.91% <15.38%> (+0.85%)	⬆️
keras-tensorflow	?
keras-torch	60.62% <30.76%> (-0.26%)	⬇️

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.

[dataCenter430]

Hi, @gbaned I hope you're doing well.
I'm a big fan of Keras and I'm going to contrbute here actively.
Could you pls review my first PR when you have a chance? 🙏 Each PR is so important for me to contribute to Keras.
Best regards