## What changes were proposed in this pull request?

This issue fixes or hides 181 Java linter errors introduced by SPARK-14987 which copied hive service code from Hive. We had better clean up these errors before releasing Spark 2.0.

- Fix UnusedImports (15 lines), RedundantModifier (14 lines), SeparatorWrap (9 lines), MethodParamPad (6 lines), FileTabCharacter (5 lines), ArrayTypeStyle (3 lines), ModifierOrder (3 lines), RedundantImport (1 line), CommentsIndentation (1 line), UpperEll (1 line), FallThrough (1 line), OneStatementPerLine (1 line), NewlineAtEndOfFile (1 line) errors.
- Ignore `LineLength` errors under `hive/service/*` (118 lines).
- Ignore `MethodName` error in `PasswdAuthenticationProvider.java` (1 line).
- Ignore `NoFinalizer` error in `ThreadWithGarbageCleanup.java` (1 line).

## How was this patch tested?

After passing Jenkins building, run `dev/lint-java` manually.
```bash
$ dev/lint-java
Checkstyle checks passed.
```

Author: Dongjoon Hyun <dongjoon@apache.org>

Closes #12831 from dongjoon-hyun/SPARK-15053.