[SPARK-12747][SQL] Use correct type name for Postgres JDBC's real array

[viirya]

https://issues.apache.org/jira/browse/SPARK-12747

Postgres JDBC driver uses "FLOAT4" or "FLOAT8" not "real".

[SparkQA]

Test build #49147 has finished for PR 10695 at commit 8f84ea4.

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

[SparkQA]

Test build #49154 has started for PR 10695 at commit 0389ea8.

[viirya]

Type mapping for various types *** FAILED *** (1 second, 877 milliseconds)
[info] Array(0.11, 0.22) did not equal Array(0.11, 0.22) (PostgresIntegrationSuite.scala:84)

Array(0.11, 0.22) did not equal Array(0.11, 0.22)?

[SparkQA]

Test build #49209 has finished for PR 10695 at commit c731a93.

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

[SparkQA]

Test build #49220 has finished for PR 10695 at commit 3a9acaa.

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

[SparkQA]

Test build #49233 has finished for PR 10695 at commit e9ec20f.

• This patch passes all tests.
• This patch merges cleanly.
• This patch adds no public classes.

[blbradley]

Looks good, but you should squash your commits into one.

[JoshRosen]

@blbradley, no need to squash the commits; that happens automatically as part of our PR merge script.

[viirya]

ping @rxin

[viirya]

ping @liancheng @yhuai @rxin Can you see if this fixing is good for you? Thanks.

[blbradley]

This is blocking me. Can we get it merged soon?

I'm waiting to submit another PR to fix DecimalType also.

[viirya]

ping @liancheng @yhuai @rxin How do you think about this? Can we merge it now?

[rxin]

Thanks - I've merged this in master and branch-1.6.