SNOW-649836: insert null into a variant column

What is the current behavior?

I have a variant column with multiple fields inside it, whenever there is a None, it doesnt insert that particular key in variant column and just skips it -

df2 = pd.DataFrame([('Mark', {'K': 2,'Z':None}), ('Luke', {'K': 4,'Z':'bla'})], columns=['ID', 'V'])

when i do write_pandas on it using this

# Write the data from the DataFrame to the table named "customers".
success, nchunks, nrows, _ = write_pandas(ctx, df2, 'customers')

This is how it inserts in snowflake now.

row   ID           V
1        Mark    { "K": 2 }
2        Luke     { "K": 4, "Z": "bla" }

What is the desired behavior?

It should look in snowflake table like this.

row   ID           V
1        Mark    { "K": 2 ,"Z":NULL}
2        Luke     { "K": 4, "Z": "bla" }

How would this improve snowflake-connector-python?

There is no option available to insert null value for a key in variant columns, snowflake allows it thought it interface, but its not possible to do it using write_pandas

References, Other Background

Start at the write_pandas entry point and reproduce the DataFrame example with a variant column containing a None value. Trace how nested dictionaries are serialized before insertion, then verify that the resulting row preserves the Z key with a NULL value while the existing non-null row remains unchanged.