Lower LightGBM test AUC base value

[simonzhaoms]

Description

The new LightGBM from SynapseML gives lower AUC value which makes smoke tests failed.
This PR lowers the test base value.

Related Issues

See #1615

Checklist:

• I have followed the contribution guidelines and code style for this project.
• I have added tests covering my contributions.
• I have updated the documentation accordingly.
• This PR is being made to staging branch and not to main branch.

[miguelgfierro]

@anargyri the change made than Ilia requires to also change the result of AUC, meaning, do we still need to merge this PR?

[miguelgfierro]

removing PR blocker

[miguelgfierro]

Approved: waiting for the decision of Ilia and Andreas

[codecov-commenter]

Codecov Report

Merging #1619 (73b6a0d) into staging (cf7bfd8) will increase coverage by 2.51%.
The diff coverage is n/a.

❗ Current head 73b6a0d differs from pull request most recent head b305bb4. Consider uploading reports for the commit b305bb4 to get more accurate results

@@             Coverage Diff             @@
##           staging    #1619      +/-   ##
===========================================
+ Coverage    56.68%   59.20%   +2.51%     
===========================================
  Files           88       88              
  Lines         8956     8956              
===========================================
+ Hits          5077     5302     +225     
+ Misses        3879     3654     -225     

Flag	Coverage Δ
pr-gate	59.20% <ø> (+2.51%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
recommenders/evaluation/spark_evaluation.py	87.05% <0.00%> (+0.44%)	⬆️
recommenders/models/sasrec/model.py	85.19% <0.00%> (+58.91%)	⬆️
recommenders/models/sasrec/sampler.py	91.66% <0.00%> (+60.41%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 735df87...b305bb4. Read the comment docs.

[anargyri]

@anargyri the change made than Ilia requires to also change the result of AUC, meaning, do we still need to merge this PR?

Yes, I think we need to merge this PR. The other PR (early stopping parameter) does not affect AUC at all.