Preview plugin: JSON Table Viewer

**Waiting for requests from scientists and a use case before implementation.**

Follow up on https://github.com/BlueBrain/nexus/issues/2565

It was mentioned by scientists that they might move from XML to JSON files for pipeline results.

The current idea would be to "flatten" the JSON; the columns of the table are all the possible "leaves" of the JSON object.

If there is one JSON, this could be a 2-column table: the first column with the leaves' path, and the second with the values.
If there are multiple JSON files, the header would be the leaves' path, and each row would be one JSON file.

Example:

```jsonld
{
"name": "bbp",
"address": {
"city" : {
"label":"Geneva",
"postcode":"1202"
},
"country": {
"label":"Switzerland",
"code":"CH"
}
}
}
```

| name | address.city.label | address.city.postcode | address.country.label | address.country.code |
|-------|------------------|-----------------------|-----------------------|-----------------------|
| bbp | Geneva | 1202 | Switzerland | CH |

| Property | Value |
|----------|-------|
| name | bbp |
| address.city.label | Geneva |
| address.city.postcode | 1202 |
| address.country.label | Switzerland |
| address.country.code | CH |

Resources:

https://github.com/Growmies/json-to-table

No files, tests, or entry points are named. Start by reviewing the linked issue 2565 and the json-to-table resource, then confirm a scientist use case and the required flattening behavior for one or multiple JSON files. Done means the requirements and table output are agreed before implementation begins.

Written by the indexing model from the issue text.