[DOCS] Define metrics

There seem to be various definitions for "metrics" in the Elastic Stack (and in particular in machine learning features). These should be clarified in our glossary and as necessary when they're used in various books.

For example:
1) The ML Data Visualizer classifies "metrics" as numeric fields:
 This use of the term seems problematic to me.

2) The ML anomaly detection job wizard refers to adding metrics, which in this case are the ML functions applied to (any type of) fields:

3) Metricbeat does not seem to define "metrics" but seems to refer to anything you can measure about a system or service; things like CPU usage, memory, file system, disk IO, network IO statistics, etc (https://www.elastic.co/beats/metricbeat).

4) The old definitive guide says "most metrics are simple mathematical operations (for example, min, mean, max, and sum) that are calculated using document values. In practical terms, metrics allow you to calculate quantities such as the average salary, or the maximum sale price, or the 95th percentile for query latency". (https://www.elastic.co/guide/en/elasticsearch/guide/2.x/_metrics.html)

5) The concept in (4) seems to be reflected in newer Elasticsearch documentation related to "metrics aggregations". There are also references to "numeric metrics aggregations" as "a special type of metrics aggregation which output numeric values"(https://www.elastic.co/guide/en/elasticsearch/reference/7.6/search-aggregations-metrics.html).

6) There is an "Elastic Metrics" product: (https://www.elastic.co/infrastructure-monitoring), which seems to align with the Metricbeat concept of metrics.

5) Online dictionaries have definitions such as these: "a standard for measuring or evaluating something; basis for assessment" or "metrics are what you measure". See also https://en.wikipedia.org/wiki/Metric