[sml] add chi2 stats in jax

[lizzy-0323]

Pull Request

What problem does this PR solve?

使用SPU实现chi2统计量的计算
Issue Number: Fixed #591

Possible side effects?

• Performance:
  Test:

OK
Chi2 stats result:
[ 10.817831    3.7107298 116.31106    67.04839  ]
Sklearn chi2 stats result:
[ 10.81782088   3.7107283  116.31261309  67.0483602 ]
P value result:
[0.00416027 0.14996627 0.         0.        ]
Sklearn p value result:
[4.47651499e-03 1.56395980e-01 5.53397228e-26 2.75824965e-15]

Emulation:

=========================
Running time in Sklearn: 0.005s
=========================
Running time in SPU: 3.628s
=========================
Chi2 stats result:
[ 10.821831    3.7121124 116.352936   67.072624 ]
Sklearn chi2 stats result:
[ 10.81782088   3.7107283  116.31261309  67.0483602 ]
P value result:
[0.00415421 0.1498375  0.         0.        ]
Sklearn p value result:
[4.47651499e-03 1.56395980e-01 5.53397228e-26 2.75824965e-15]
[2024-04-25 12:54:47,659] Shutdown multiprocess cluster...

• Backward compatibility:

[deadlywing]

另外，，univariate_selection.py 这个文件缺少lisence

[deadlywing]

以及，bazel文件的格式不对，，以下两个文件应该是缺少空行，您可以在本地用buildifier格式化后再push
sml/feature_selection/BUILD.bazel # reformat
sml/feature_selection/tests/BUILD.bazel # reformat

[lizzy-0323]

已按照修改意见修改。

[deadlywing]

LGTM