Updates to data/metadata following application to 15m

[sgreenbury]

Following the use of popgetter to a-b-street/15m#10, opening this issue to capture changes to the data/metadata that would improve usability.

• Ensure source_metric_id for every metric is either the census table name that it is derived from or the original metric ID. Currently the values are:

  In [5]: df["metric_source_id"].value_counts()
  Out[5]:
  shape: (3, 2)
  ┌──────────────────┬────────┐
  │ metric_source_id ┆ count  │
  │ ---              ┆ ---    │
  │ str              ┆ u32    │
  ╞══════════════════╪════════╡
  │ TBD              ┆ 434536 │
  │ MS_POPULATION    ┆ 1364   │
  │ TOTAL            ┆ 1      │
  └──────────────────┴────────┘

This will enable the CLI search: popgetter metrics --source-metric-id <SOURCE_METRIC_ID>

• Aim for the same derived metrics to have the same metric metadata (human readable name, hxltag etc) for different countries
• Add metadata field to indicate if derived metric or not. This could also be included in the text instead but might be clearer if it has its own field enum
• Add a measure for the resolution of the geometry level to the metadata. This could be the approximate number of individuals for the average zone of the resolution or a country-specific ranking (e.g. 1 for smallest area, 2 for the second smallest, etc)
• Add the coordinate reference system for a geometry as metadata. This is captured in Popgetter Coordinate Reference System strategy #142.
• Add metadata sufficient to enable Help the user generate attribution text popgetter-cli#83