[MINOR][SQL]Change ThreadLocal.withInitial to Scala lambda syntax

[10110346]

What changes were proposed in this pull request?

Currently, Scala has updated to Scala 2.12.8 , so we can change this to Scala lambda syntax

How was this patch tested?

Existing tests.
And Manual tested:

== Physical Plan ==
*(3) HashAggregate(keys=[attribute#237L], functions=[sum(cnt#227L)])
+- Exchange hashpartitioning(attribute#237L, 5)
   +- *(2) HashAggregate(keys=[attribute#237L], functions=[partial_sum(cnt#227L)])
      +- *(2) HashAggregate(keys=[nested#223.attribute#243L], functions=[count(1)])
         +- Exchange hashpartitioning(nested#223.attribute#243L, 5)
            +- *(1) HashAggregate(keys=[nested#223.attribute AS nested#223.attribute#243L], functions=[partial_count(1)])
               +- *(1) Project [nested#223]
                  +- *(1) Scan ExistingRDD[nested#223,value#224L]

[SparkQA]

Test build #100003 has finished for PR 23295 at commit 361ed5c.

• This patch fails due to an unknown error code, -9.
• This patch merges cleanly.
• This patch adds no public classes.

[SparkQA]

Test build #100014 has finished for PR 23295 at commit 361ed5c.

• This patch fails SparkR unit tests.
• This patch merges cleanly.
• This patch adds no public classes.

[10110346]

retest this please

[SparkQA]

Test build #100059 has finished for PR 23295 at commit 361ed5c.

• This patch fails SparkR unit tests.
• This patch merges cleanly.
• This patch adds no public classes.

[10110346]

retest this please

[SparkQA]

Test build #100081 has finished for PR 23295 at commit 361ed5c.

• This patch fails PySpark unit tests.
• This patch merges cleanly.
• This patch adds no public classes.

[srowen]

We still build against 2.11 too. Unless you have built with 2.11 and this works, we can't merge this.

[10110346]

The ThreadLocal.withInitial lambda syntax is supported in java1.8, and this is already used in ReadAheadInputStream class.

[srowen]

This isn't Java code though. It may well work, but it needs to be tested against 2.11. The pull request builder doesn't test this. Have you? I remember some odd issues with Scala's willingness to match closures to single-arg function args, and given this comment, I think we need to be sure first.

[10110346]

I see ,I have a mistake, this is scala code, it can not work for scala 2.11
thanks @srowen, I will close this PR