SNOW-1360263: [Local Testing] Tests for several issues.

[sfc-gh-jrose]

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

   Fixes SNOW-1360263
   Fixes SNOW-1668862
   Fixes SNOW-1694649
   Fixes SNOW-1707286

2. Fill out the following pre-review checklist:

   • I am adding a new automated test(s) to verify correctness of my new code
     • If this test skips Local Testing mode, I'm requesting review from @snowflakedb/local-testing
   • I am adding new logging messages
   • I am adding a new telemetry message
   • I am adding new credentials
   • I am adding a new dependency
   • If this is a new feature/behavior, I'm adding the Local Testing parity changes.

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

This PR adds tests to verify fixes proposed by community member @tvdboom.

Notes from their PR #2373:

SNOW-1668981

The code block

for k, v in expr_to_alias.items():
if v == exp.child.name:
expr_to_alias[k] = quoted_name

wrongly overwrites the expr_to_alias of the columns in dataframe anti (see minimal reproducible example in #2305). I am unsure when this code block would be desirable. Tests pass without them. Still, better for a maintainer to have a good look at this change. If it's required, we could make it optional under a condition yet to be determined.

SNOW-1360263

The windows variable needs to be ordered by the order of the index in res_index.

SNOW-1668862

Literal columns were always created with nullable=False, which made lit(None) columns fail. Now, lit(None) returns a nullable column.

SNOW-1694649

In the merge statement, filtering on the join_condition resulted in a dataframe with unordered indices. This resulted in a failure when calling new_column[either_isna]. Reseting the index solves the issue.

SNOW-1707286

Can't operate on None and datetime. Return None directly.

jrose Notes:

SNOW-1668981: I've removed the fix for this from the original PR as it appears to also affect live mode and will need a larger change.

SNOW-1360263: The proposed change broke in some edge cases when I ran tests against it. I've amended this part of the change to extract the ordering from the rolling window operation instead.

SNOW-1668862, SNOW-1694649, SNOW-1707286: All look good as is.

[sfc-gh-aling]

shall we validate the table after saving?

[sfc-gh-jrose]

Probably not, will update.