Add row highlighting support in the table

[JoshuaJewell]

Problem

Users need a way to visually mark or identify rows of interest while filtering and scanning data (e.g., "What samples contain reads for Seq42?").

Proposed behaviour

Add support for highlighted rows in the table.

Behaviour should support at least one of these flows:

1. Highlight rows that match a user-defined condition.
2. Highlight rows matching the current global filter/search context.
3. Allow the user to manually highlight rows.

Distinction from existing filtering strategy:

• highlighting only marks rows visually
• highlighted rows should not automatically change column totals unless the highlight is also used as an active filter
• highlighting must allow for hiding of columns summing to 0 in active filter mode (hide samples with no reads under row filter)

Suggested behaviour

• Highlighted rows receive a clear visual treatment.
• Highlight state should behave predictably when filtering.
• If a highlighted row is hidden by a filter, it should reappear highlighted when the filter is cleared.

Open questions

• Should highlights be temporary UI state or saved with the analysis/session?
• Should multiple highlight categories/colours be supported?