Fix DeepFM dense-only loading, DNN save tracking, and FNN logits

[shenweichen]

Summary

This PR addresses the first batch of triaged open issues:

• Fixes Implementing fix from Issue#344 #529 by constructing DNN activation layers without mutating a tracked list after creation.
• Fixes Error while loading models with only dense features #508 by avoiding an empty FM branch in dense-only DeepFM models.
• Fixes FNN model is not implemented yet... #459 by making FNN use only the DNN logit instead of adding a WDL-style linear logit.

It also updates Linear.add_weight calls to pass name= explicitly, which keeps the touched DeepFM dense-only path compatible with modern Keras.

Validation

• python -m pytest tests/models/DeepFM_test.py::test_DeepFM_dense_only_model_io tests/layers/core_test.py::test_DNN_output_activation tests/models/FNN_test.py::test_FNN_does_not_add_wide_linear_logit -q
• python -m pytest tests/feature_test.py::test_create_embedding_matrix_reuses_same_embedding_name tests/feature_test.py::test_create_embedding_matrix_rejects_inconsistent_shared_embedding -q
• Manual Hash(4, mask_zero=True) range check

[codacy-production]

Up to standards ✅

🟢 Issues 0 issues

Results:
0 new issues

View in Codacy

🟢 Metrics 4 complexity · -5 duplication

Metric	Results
Complexity	4
Duplication	-5

View in Codacy

_{NEW Get contextual insights on your PRs based on Codacy's metrics, along with PR and Jira context, without leaving GitHub. Enable AI reviewer
TIP This summary will be updated as you push new changes.}

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 78.31%. Comparing base (a827877) to head (930367f).
⚠️ Report is 1 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #555      +/-   ##
==========================================
+ Coverage   78.23%   78.31%   +0.08%     
==========================================
  Files          61       61              
  Lines        3331     3325       -6     
==========================================
- Hits         2606     2604       -2     
+ Misses        725      721       -4     

Flag	Coverage Δ
pytest	78.31% <100.00%> (+0.08%)	⬆️

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.