[Fix #9939] Fix/hash as last array item

[martinjaimem]

Fix autocorrect bug with HashAsLastArrayItem

Fixes #9939

When running autocorrect with:

Style/TrailingCommaInHashLiteral:
  EnforcedStyleForMultiline: comma

Style/TrailingCommaInArrayLiteral:
  EnforcedStyleForMultiline: comma

Style/HashAsLastArrayItem:
  EnforcedStyle: no_braces

there was an edge case where it created incorrect code:

before autocorrect

x = [
  1,
  {
    a: 1,
    b: 2,
  },
]

after autocorrect:

x = [
  1,
  
    a: 1,
    b: 2,
  ,
]

With this PR, the result of applying autocorrect makes executable code =>

x = [
  1,
  
    a: 1,
    b: 2,

]

Notes:

• The PR contains commits that are not squashed, that's because I specifically wanted to receive feedback before and then I can squash the two commits (this way I can show the two approaches)
• The first commit represents the original approach that only corrected the code within the node, it worked fine with a one-line array but generated the wrong code with multi-line arrays. Maybe it was okay to leave it that way but It was better to mark it as unsafe, some feedback would be awesome.
• The second commit corrects the code within the node and its parent (I don't know if that's a good practice, I'm really open to some feedback)

---

Before submitting the PR make sure the following are checked:

• The PR relates to only one subject with a clear title and description in grammatically correct, complete sentences.
• Wrote good commit messages.
• Commit message starts with [Fix #issue-number] (if the related issue exists).
• Feature branch is up-to-date with master (if not - rebase it).
• Squashed-related commits together.
• Added tests.
• Ran bundle exec rake default. It executes all tests and runs RuboCop on its own code.
• Added an entry (file) to the changelog folder named {change_type}_{change_description}.md if the new code introduces user-observable changes. See changelog entry format for details.

[thearjunmdas]

Should there be a testcase for this scenario as well?

 [
    1,
    2,
    {
      one: 1,
      two: 2,
    },
    3,
    4
]

p.s I am not an official maintainer. Randomly came across your PR so I thought I could take a look to understand the codebase better..

[martinjaimem]

I added that scenario 👍 good call

[bbatsov]

The proposed fix looks good to me. Thanks!