[SPARK-41779][SPARK-41771][CONNECT][PYTHON] Make __getitem__ support filter and select

[zhengruifeng]

What changes were proposed in this pull request?

Make dataframe __getitem__ support:
1, filter: cdf[cdf.a.isin(1, 2, 3)]
2, select: cdf[["col1", cdf.a]]
3, index: cdf[0]

Why are the changes needed?

to be consistent with PySpark

Does this PR introduce any user-facing change?

yes

How was this patch tested?

added UT

[zhengruifeng]

this pr is to fix

File "/.../spark/python/pyspark/sql/connect/column.py", line 364, in pyspark.sql.connect.column.Column.isin
Failed example:
    df[df.name.isin("Bob", "Mike")].collect()
Exception raised:
    Traceback (most recent call last):
      File "/.../miniconda3/envs/python3.9/lib/python3.9/doctest.py", line 1336, in __run
        exec(compile(example.source, filename, "single",
      File "<doctest pyspark.sql.connect.column.Column.isin[1]>", line 1, in <module>
        df[df.name.isin("Bob", "Mike")].collect()
      File "/.../workspace/forked/spark/python/pyspark/sql/connect/dataframe.py", line 888, in __getitem__
        return col(name)
      File "/.../workspace/forked/spark/python/pyspark/sql/connect/functions.py", line 161, in col
        return Column(ColumnReference(col))
      File "/.../spark/python/pyspark/sql/connect/expressions.py", line 322, in __init__
        assert isinstance(name, str)
    AssertionError

[zhengruifeng]

cc @HyukjinKwon @grundprinzip

[zhengruifeng]

thanks, merged into master