SNOW-869536 Fix buggy behavior in DataFrame.to_local_iterator

[sfc-gh-stan]

Please answer these questions before submitting your pull requests. Thanks!

1. What GitHub issue is this PR addressing? Make sure that there is an accompanying issue to your PR.

   Fixes SNOW-869536: Iterators from to_local_iterator stop returning results after another query occurs #945

2. Fill out the following pre-review checklist:

   • I am adding a new automated test(s) to verify correctness of my new code
   • I am adding new logging messages
   • I am adding a new telemetry message
   • I am adding new credentials
   • I am adding a new dependency

3. Please describe how your code solves the related issue.

   Issue: All queries issued from a Snowpark Session object are executed using the same SnowflakeCursor instance, this causes unexpected, wrong behavior when we fetch results from an iterator but have code that executes a different query half way, e.g.

    df = session.create_dataframe([[1,2,3],[4,5,6]], schema=["a", "b", "c"]) 
    my_iter = df.to_local_iterator()
    row_counter
    for row in my_iter:
       len(df.schema.fields)  # this executes a schema query, which overwrites properties of the cursor object
       row_counter += 1
    print(counter)  # this prints 1 since schema query only returns 1 row, but should be 2
   

   Fix: This PR changes src/snowflake/snowpark/_internal/server_connection.py create a new cursor object locally to read data from the last executed query and continues to use this cursor object in the iterator to achieve isolation between the iterator and the following queries executed within the session.

[sfc-gh-stan]

to_pandas doesn't have this issue, SnowflakeCursor.fetch_pandas_batches already handles the isolation.

[sfc-gh-sfan]

Can we put this into comment instead? This seems more informative than "fix for" :)

[sfc-gh-sfan]

Did we remove this because the param is unused?

[sfc-gh-stan]

Yes

[sfc-gh-sfan]

Can we put this into comment instead? This seems more informative than "fix for" :)