[GEN-1516, GEN-1555, GEN-1576] make table update cohort specific

[danlu1]

Problem:

1. Currently, the update table step is NOT cohort-specific which means that all the data in the internal tables get overwritten.
2. Outputs are either saved to prod environment or local. We would like to be able to save them in staging project for testing or validation purpose.
3. Need Docker image for the feature branch specific for testing
4. Can't test this data processing step individually in Tower.
5. None values in the Synapse table has been converted to blank since None is in default na_values list of pandas dataframe.
6. Non-informative backslash in the input data.

Solution:

1. Table queries are updated to include cohort filter.
2. Wipe table line has been updated to wipe cohort lines only.
3. Staging projects are added and Staging tables are used.
4. The table_update step has been added in the Nextflow workflow. Nextflow scripts for table_update step can intake docker image as a parameter.
5. Remove None from the na_values of the asDataFrame() function.
6. A new column specific remove_backslash function is added to remove non-informative backslashes in the data before pushing to Synapse table.

Testing:

1. Unit tests for new and modified functions are added
2. Comparison scrips were run to compare versions of the tables and got expected results.
3. Scripts run through both in EC2 instance and Tower.

[rxu17]

LGTM! Just minor comments

[thomasyu888]

🔥 LGTM! Please make sure all of Rixing's comments are addressed!