[Data Table] Mitigate column data bloat to allow for better visualization

(Pulled from associated UDN)

Data Table quality of life suggestion
Data tables are one of the more useful asset types for managing 'big data' in the engine. Problem is, the spreadsheet view, which is one of the most useful aspects of it, is deeply inadequate for presenting data in that spread sheet view in a useful manner.

By turning all the fields in the record into json structures, it adds noisy fluff to the spreadsheet view that in many cases makes the spreadsheet view nearly unreadable, and/or the json structure of the field bloats up the column data so much, that visualizing that column is effectively useless.

Here's an example:
[Link Removed]

This may be too "hacky" to adopt as an engine solution to this problem, but I show it to spark consideration for something that solves this issue.

With a minimal change to engine code, I've added a project level configurable hook to format a column at the project level, instead of doing the default json-ification stuff.

In DataTableUtils.cpp
[Link Removed]

Through these callbacks, the project level can completely customize that data table representation/summary of anything type they wish.

Here is the same table again, with this customization enabled. The great thing about this, is that you can customize the data table summary view of fields down to the individual struct type for example.
[Link Removed]

You can see by comparison how this customization can be used to present the data in an entirely more useful way.

I'm happy to contribute this tiny little engine level hook that facilitates this, but I mainly wanted to bring attention to this gaping inadequacy of the data table view, and one very easy code side way this can take a huge step forward for readability and usefulness.

Thoughts?