Initial entry point to data generation for scale test

[wjxiz1992]

As titled.

close #8813

This PR aims to provide the initial entry point to the data generation application for scale test.
The design and user interface are described at #8813 (comment)

still DRAFT version, posted for early review and feedbacks.

One example command to test it locally:

$SPARK_HOME/bin/spark-submit \
--master spark://*:7077 \
--conf spark.driver.memory=10G \
--conf spark.executor.memory=32G \
--conf spark.sql.parquet.int96RebaseModeInWrite=CORRECTED \
--conf spark.sql.parquet.datetimeRebaseModeInWrite=CORRECTED \
--class com.nvidia.rapids.tests.scaletest.ScaleTestDataGen \
--jars $SPARK_HOME/examples/jars/scopt_2.12-3.7.1.jar \
./target/datagen_2.12-23.10.0-SNAPSHOT-spark332.jar \
1 \
10 \
parquet \
file:/*/testdata

Giving an example to show the actual disk size the data will take so we have basic impression:
For Scale=1, Complexity=1 and parquet file:

2.2M    a_facts
115M    b_data
28M     c_data
26M     d_data
282M    e_data
296K    f_facts
150M    g_data

For Scale=1, Complexity=10 and parquet file:

2.7M    a_facts
302M    b_data
21M     c_data
57M     d_data
295M    e_data
584K    f_facts
150M    g_data

For Scale=10, Complexity=10 and parquet file:

27M     a_facts
3.0G    b_data
295M    c_data
655M    d_data
2.9G    e_data
4.7M    f_facts
1.5G    g_data

• Basic code structure
• README doc for how to use it
• Tests for data and scale queries

[wjxiz1992]

resolve comments and add CorrelatedKeyGroup for key groups in tables.

[revans2]

Mostly looks good. Just a few nits

[wjxiz1992]

resolve more comments.

[revans2]

Thanks for your patience in doing the rework. I think we are really close now and it looks really good.

Thanks

[wjxiz1992]

Thanks for review! add a flag arg "--overwrite" and resolve the rest comments.

[revans2]

ScalaTest.md appears to also have some TODOs in it. Are there plans to fix that?

[wjxiz1992]

it. Are there plans to fix that?

Yes, in this PR we only add the data generation part, but not include the queries to run with them.

I put a TODO part in Test Query Sets sections to deal with the up coming work for #8814 and #8816 so that this scale tset is a complete integral test suite.

[revans2]

build