Add description for custom metrics tags

[R-HNF]

What does this PR do?

• Add description for custom metrics tags. This tags is the field that can be configured under custom_queries.
• Also, made some minor fixes to the README.md.

Motivation

There is no description for custom metrics tags in README.md. I think it's necessary to clarify that, unlike the tags configured under instances, the tags configured under custom_queries apply only to custom metrics.
Please refer the following:

integrations-core/elastic/datadog_checks/elastic/elastic.py

Lines 510 to 612 in e7589a6

	def _process_custom_metric(
	self,
	value,
	data_path,
	value_path,
	dynamic_tags,
	xtype,
	metric_name,
	tags=None,
	):
	"""
	value: JSON payload to traverse
	data_path: path to data right before metric value or right before list of metric values
	value_path: path to data after data_path to metric value
	dynamic_tags: list of dynamic tags and their value_paths
	xtype: datadog metric type, default to gauge
	metric_name: datadog metric name
	tags: list of tags that should be included with each metric submitted
	"""
	tags_to_submit = deepcopy(tags)
	path = '{}.{}'.format(data_path, value_path)
	# Collect the value of tags first, and then append to tags_to_submit
	for dynamic_tag_path, dynamic_tag_name in dynamic_tags:
	# Traverse down the tree to find the tag value
	dynamic_tag_value = get_value_from_path(value, dynamic_tag_path)
	# If tag is there, then add it to list of tags to submit
	if dynamic_tag_value is not None:
	dynamic_tag = '{}:{}'.format(dynamic_tag_name, dynamic_tag_value)
	tags_to_submit.append(dynamic_tag)
	else:
	self.log.debug("Dynamic tag is null: %s -> %s", path, dynamic_tag_name)
	# Now do the same for the actual metric
	branch_value = get_value_from_path(value, value_path)
	if branch_value is not None:
	if xtype == "gauge":
	self.gauge(metric_name, branch_value, tags=tags_to_submit)
	elif xtype == "monotonic_count":
	self.monotonic_count(metric_name, branch_value, tags=tags_to_submit)
	elif xtype == "rate":
	self.rate(metric_name, branch_value, tags=tags_to_submit)
	else:
	self.log.warning(
	"Metric type of %s is not gauge, monotonic_count, or rate; skipping this metric", metric_name
	)
	else:
	self.log.debug("Metric not found: %s -> %s", path, metric_name)
	def _run_custom_queries(self, admin_forwarder, base_tags):
	self.log.debug("Running custom queries")
	custom_queries = self._config.custom_queries
	for query_endpoint in custom_queries:
	try:
	columns = query_endpoint.get('columns', [])
	data_path = query_endpoint.get('data_path')
	raw_endpoint = query_endpoint.get('endpoint')
	static_tags = query_endpoint.get('tags', [])
	payload = query_endpoint.get('payload', {})
	endpoint = self._join_url(raw_endpoint, admin_forwarder)
	data = self._get_data(endpoint, data=payload)
	# If there are tags, add the tag path to list of paths to evaluate while processing metric
	dynamic_tags = get_dynamic_tags(columns)
	tags = base_tags + static_tags
	# Traverse the nested dictionaries to the data_path and get the remainder JSON response
	value = get_value_from_path(data, data_path)
	for column in columns:
	metric_type = column.get('type', 'gauge')
	# Skip tags since already processed
	if metric_type == 'tag':
	continue
	name = column.get('name')
	value_path = column.get('value_path')
	if name and value_path:
	# At this point, there may be multiple branches of value_paths.
	# If value is a list, go through each entry
	if isinstance(value, list):
	value = value
	else:
	value = [value]
	for branch in value:
	self._process_custom_metric(
	value=branch,
	data_path=data_path,
	value_path=value_path,
	dynamic_tags=dynamic_tags,
	xtype=metric_type,
	metric_name=name,
	tags=tags,
	)
	except Exception as e:
	self.log.error("Custom query %s failed: %s", query_endpoint, e)
	continue

Minor fixes:

• Remove a trailing space.
• Adjust whitespace in curl commands.
• Use an absolute URL for the link.
  • To prevent redirection to the following URL when accessed from GitHub:
    https://github.com/DataDog/integrations-core/blob/master/account/settings/agent/latest

Review checklist (to be filled by reviewers)

• Feature or bugfix MUST have appropriate tests (unit, integration, e2e)
• Add the qa/skip-qa label if the PR doesn't need to be tested during QA.
• If you need to backport this PR to another branch, you can add the backport/<branch-name> label to the PR and it will automatically open a backport PR once this one is merged

[R-HNF]

Reference:

integrations-core/bitwarden/README.md

Line 37 in e7589a6

[2]: https://app.datadoghq.com/account/settings/agent/latest

[aliciascott]

looks good