Add dask module annotation

[lazargugleta]

Description

Dask library is based on pandas classes and methods for data frames with the extension for large parallel computing and shuffling. Some module-specific methods required annotating, including file system methods from pandas such as read_csv.

Type of change

• [x] New feature (non-breaking change which adds functionality)

How Has This Been Tested?

Before adding annotation:

import dask.dataframe as dd
df = dd.read_csv("tests/simple_data.csv")

After:

import dask.dataframe as dd
df = dd.read_csv("tests/simple_data.csv")
df.pop("a")

[lazargugleta]

I see, you are correct @lionsardesai. I mistook those two for a method of DataFrame class. Would explode, melt, pop, drop, dropna, drop_duplicates be a part of it since they mutate the object? What side_effects should visualize and head have If those should even have annotation?

[lazargugleta]

Thanks to @lionsardesai for giving me better insight into what annotations are required for dask. After checking all methods in detail, the only method that could be used for annotations is pop because it changes the variable inplace. I updated the annotations and test files.

[lazargugleta]

Very nice! 🎉 Thanks to you too @lionsardesai