Price stats API: Allow stats to be grouped by specific field

[TTalex]

Problem

There are currently no way of fetching stats grouped by a value other than pulling the entire database.

For example, it would be great to be able to fetch price counts per users, or average price per location, or price count per day, and so on...

In the upcoming challenge page, this would help

• Rank users by price count (group by users)
• Display number of contributions per day on a graph (group by date)

Proposed solution

Add a new GET /api/v1/prices/grouped_stats endpoint, with mandatory param group_by matching an existing field.
For example, GET /api/v1/prices/grouped_stats?group_by=date&size=2 would give

{
  "items": [
    {
      "date": "2018-03-27",
      "price__count": 24,
      "price__min": 0.47,
      "price__max": 7.8,
      "price__avg": 2.55,
      "price__sum": 61.2
    },
    {
      "date": "2018-03-31",
      "price__count": 13,
      "price__min": 0.54,
      "price__max": 6.49,
      "price__avg": 3.19,
      "price__sum": 41.49
    }
  ],
  "page": 1,
  "pages": 55,
  "size": 2,
  "total": 110
}

By default, results are ordered following the group_by parameter (here by date), but we could also add an optional param order_by to sort the result with one of the computed result.
For example, GET /api/v1/prices/grouped_stats?group_by=date&size=2&order_by=-price__count would give

{
  "items": [
    {
      "date": "2021-01-11",
      "price__count": 37,
      "price__min": 0.4,
      "price__max": 5.99,
      "price__avg": 2.05,
      "price__sum": 75.8
    },
    {
      "date": "2018-08-09",
      "price__count": 37,
      "price__min": 0.59,
      "price__max": 12.15,
      "price__avg": 2.67,
      "price__sum": 98.71
    }
  ],
  "page": 1,
  "pages": 55,
  "size": 2,
  "total": 110
}