Support for Elasticsearch version 7.13.0
Adds support for compatibility header for Elasticsearch. If the environment variable ELASTIC_CLIENT_APIVERSIONING is set to true or 1, the client will send the headers Accept and Content-Type with the following value: application/vnd.elasticsearch+json;compatible-with=7.
Better detection of Elasticsearch and Enterprise Search clients in the meta header used by cloud.

The REST API tests now use an artifact downloaded from the Elastic servers instead of depending of cloning elasticsearch locally. Check the README for more information.
New parameter include_unloaded_segments in cat.nodes, nodes.stats: If set to true segment stats will include stats for segments that are not currently loaded into memory
New parameter summary in ingest.get_pipeline: Return pipelines without their definitions (default: false)
New parameter index_details in snapshot.get: Whether to include details of each index in the snapshot, if those details are available. Defaults to false.
New endpoint features.reset_features, ingest/geo_ip_stats
New experimental endpoints: shutdown.delete_node, shutdown.get_node, shutdown.put_node.

Refactored test tasks, made it easier to run the tests by default.
New experimental endpoints: fleet.global_checkpoints, searchable_snapshots.cache_stats.
New beta endpoints: security.clear_cached_service_tokens, security.create_service_token, security.delete_service_token, security.get_service_accounts, security.get_service_credentials
New endpoints: machine_learning.delete_trained_model_alias, machine_learning.preview_data_frame_analytics, machine_learning.put_trained_model_alias.
APIs migrated from experimental or beta to stable: machine_learning.delete_data_frame_analytics, machine_learning.delete_trained_model, machine_learning.estimate_model_memory, machine_learning.explain_data_frame_analytics, machine_learning.get_data_frame_analytics, machine_learning.get_data_frame_analytics_stats, machine_learning.get_trained_models, machine_learning.get_trained_models_stats, machine_learning.put_data_frame_analytics, machine_learning.put_trained_model, machine_learning.start_data_frame_analytics, machine_learning.stop_data_frame_analytics, machine_learning.update_data_frame_analytics
New parameter body in machine_learning.preview_datafeed: The datafeed config and job config with which to execute the preview.