[SPARK-47713][SQL][CONNECT] Fix a self-join failure

[zhengruifeng]

What changes were proposed in this pull request?

update the logic to resolve column in spark connect

Why are the changes needed?

        df = spark.createDataFrame([(1, 2), (3, 4)], schema=["a", "b"])
        df2 = df.select(df.a.alias("aa"), df.b)
        df3 = df2.join(df, df2.b == df.b)

AnalysisException: [AMBIGUOUS_COLUMN_REFERENCE] Column "b" is ambiguous. It's because you joined several DataFrame together, and some of these DataFrames are the same.
This column points to one of the DataFrames but Spark is unable to figure out which one.
Please alias the DataFrames with different names via `DataFrame.alias` before joining them,
and specify the column using qualified name, e.g. `df.alias("a").join(df.alias("b"), col("a.id") > col("b.id"))`. SQLSTATE: 42702

Does this PR introduce any user-facing change?

yes, above query can run successfully after this PR

This PR only affects Spark Connect, won't affect Classic Spark.

How was this patch tested?

added tests

Was this patch authored or co-authored using generative AI tooling?

no

[cloud-fan]

when this condition can be false if resolved.isEmpty is true?

[zhengruifeng]

I think there are two cases:

1, analyzer rules supporting missing column resolution

https://github.com/apache/spark/blob/923f04606fe6bee5913f8fce7aaa643984f79756/sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/ColumnResolutionHelper.scala#L583-L607

2, plan id missed in some way (should be bugs)

[cloud-fan]

Shall we still have the matched flag? The new code is confusing and I can't understand it even after reading your comment.

[xinrong-meng]

A comment in code would be very helpful

[zhengruifeng]

ok, let me restore matched

[zhengruifeng]

Row constructor doesn't accept duplicated names, so test in this way

[cloud-fan]

thanks, merging to master!