feat: Proper Handling of Pandoc Multiline/Grid Tables

**What problem does this solve?**
Currently, when using multiline or grid tables in Pandoc or Quarto Markdown, various Harper linting errors may show up incorrectly, such as `UnclosedQuotes`, or `LongSentences`.

**Proposed Solution**
Harper should know to treat different table cells as separate pieces of text.

**Examples**
Incorrectly gives `LongSentences` lint:

```qmd
+----------------------------+----------------------+
| This is the first cell of | This is the second |
| the table; the sentence is | cell of the table; |
| not too long, but is when | an independent |
| Harper incorrectly treats | sentence that should |
| the second cell as part of | not be considered |
| it. | part of cell 1. |
+----------------------------+----------------------+
```

**Component**
- [x] Core engine
- [ ] Plugin/Extension
- [ ] Other: _____

**Additional Context**
Documentation about different types of Pandoc/Quarto tables can be found here:

No source files or tests are named. Start by locating the core engine's handling of Pandoc or Quarto Markdown tables and compare its treatment of multiline and grid-table cells with the examples. Done means separate cells are treated as independent text so valid content does not trigger false UnclosedQuotes or LongSentences lints.

Written by the indexing model from the issue text.