Feature: CSV parsing for hub.auto

[dhiganthrao]

The purpose of this PR is to add code for auto parsing of CSV files to hub.auto. Kindly go through it and let me know where I can improve!

[github-actions]

Locust summary

Git references

Initial: e3f1c84
Terminal: f2d7ae4

---

hub/auto/tabular/csv.py

Changes:

• Name: data_from_csv
  Type: function
  Changed lines: 41
  Total lines: 41
  Changes:
  • Name: data_from_csv.csv_data
    Type: function
    Changed lines: 5
    Total lines: 5

hub/auto/infer.py

Changes:

• Name: _find_root
  Type: function
  Changed lines: 2
  Total lines: 31

hub/auto/tests/test_tabular_data.py

Changes:

• Name: assert_conversion
  Type: function
  Changed lines: 31
  Total lines: 31
• Name: test_class_sample_single_csv
  Type: function
  Changed lines: 5
  Total lines: 5
• Name: test_class_sample_multiple_csv
  Type: function
  Changed lines: 5
  Total lines: 5

[dhiganthrao]

Apologies for the lint error. Will rectify it ASAP

[mynameisvinn]

copying @McCrearyD

[verbose-void]

Apologies for the lint error. Will rectify it ASAP

no need to apologize, lint errors happen :) -- you can refer to the contributing doc for black setup with vscode.

[verbose-void]

i made some minor requests, after those are done should be ready to merge :P

[verbose-void]

last changes

[dhiganthrao]

@McCrearyD pushed the changes you've requested! Re-requesting review from you.

[verbose-void]

perfect!

[verbose-void]

CLCI was failing tests, but those were dependency issues. addressed those & got them passing. awaiting CLCI run, then merging

[verbose-void]

my previous approval was wrong, please do these changes

[dhiganthrao]

@McCrearyD made some changes with respect to your last review. Hope this is fine!

Detailing the changes I've made:

• I've added new csv files with different and random values, to check if the parser still works properly on it. Please let me know if I should change the csv files if they are unsatisfactory.
• I'm now reading a Pandas Dataframe from the csv file directly to compare it with the dataset generated by hub.auto. This makes it easier for me to check whether the column names and their data types are the same and so on.
• I've also written different test conditions to implement what you've asked me to do in your last review.

Please go through it and let me know where I can improve.

[verbose-void]

this is perfect. stellar work!