[SPARK-27459][SQL] Revise the exception message of schema inference failure in file source V2

## What changes were proposed in this pull request?

Since https://github.com/apache/spark/pull/23383/files#diff-db4a140579c1ac4b1dbec7fe5057eecaR36, the exception message of schema inference failure in file source V2 is `tableName`, which is equivalent to `shortName + path`.

While in file source V1, the message is `Unable to infer schema from ORC/CSV/JSON...`.
We should make the message in V2 consistent with V1, so that in the future migration the related test cases don't need to be modified. apache#24058 (review)

## How was this patch tested?

Revert the modified unit test cases in https://github.com/apache/spark/pull/24005/files#diff-b9ddfbc9be8d83ecf100b3b8ff9610b9R431 and https://github.com/apache/spark/pull/23383/files#diff-9ab56940ee5a53f2bb81e3c008653362R577, and test with them.

Closes apache#24369 from gengliangwang/reviseInferSchemaMessage.

Authored-by: Gengliang Wang <[email protected]>
Signed-off-by: Wenchen Fan <[email protected]>

Author: gengliangwang

sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/v2/FileTable.scala

@@ -54,7 +54,7 @@ abstract class FileTable(
     inferSchema(fileIndex.allFiles())
   }.getOrElse {
     throw new AnalysisException(
-      s"Unable to infer schema for $name. It must be specified manually.")
+      s"Unable to infer schema for $formatName. It must be specified manually.")
   }.asNullable
 
   override lazy val schema: StructType = {

sql/core/src/test/scala/org/apache/spark/sql/execution/datasources/orc/OrcQuerySuite.scala

@@ -580,7 +580,7 @@ abstract class OrcQueryTest extends OrcTest {
       val m1 = intercept[AnalysisException] {
         testAllCorruptFiles()
       }.getMessage
-      assert(m1.contains("Unable to infer schema"))
+      assert(m1.contains("Unable to infer schema for ORC"))
       testAllCorruptFilesWithoutSchemaInfer()
     }
 

sql/core/src/test/scala/org/apache/spark/sql/test/DataFrameReaderWriterSuite.scala

@@ -428,7 +428,7 @@ class DataFrameReaderWriterSuite extends QueryTest with SharedSQLContext with Be
     val message = intercept[AnalysisException] {
       testRead(spark.read.csv(), Seq.empty, schema)
     }.getMessage
-    assert(message.toLowerCase(Locale.ROOT).contains("unable to infer schema for csv"))
+    assert(message.contains("Unable to infer schema for CSV. It must be specified manually."))
 
     testRead(spark.read.csv(dir), data, schema)
     testRead(spark.read.csv(dir, dir), data ++ data, schema)