Add query_log_usage table and materialized view

What does this MR do and why?

Add query_log_usage table and materialized view where fully instrumented query metrics will be collected.

Related to: https://gitlab.com/groups/gitlab-com/gl-infra/finops/-/work_items/129+

Column	Description
event_time	Query finish time, UTC. Partition key and leading sort key.
query_id	ClickHouse query UUID. Second sort key.
root_namespace_id	Root namespace, from log_comment. Rows without one are dropped.
organization_id	Organization, from log_comment. 0 when absent.
user_id	User who triggered the query. 0 when absent.
correlation_id	Request correlation id. Joins back to the Rails logs.
application	web, sidekiq, console or test.
feature_category	GitLab feature category. '' when absent.
query_kind	Select, Insert, Create, Drop, etc.
databases	Databases the query touched.
tables	Tables touched.
query_duration_ms	Wall-clock duration.
read_rows	Rows scanned.
read_bytes	Bytes scanned.
memory_usage_bytes	Peak memory.
os_cpu_virtual_time_us	CPU time, from ProfileEvents['OSCPUVirtualTimeMicroseconds'].
exception_code	0 on success, else the ClickHouse error code.
log_comment	The raw comment, kept for keys we don't extract yet.
vcpu_seconds	ALIAS. os_cpu_virtual_time_us / 1e6.
memory_gb_seconds	ALIAS. memory_usage_bytes * query_duration_ms / 1e12. Upper bound.

References

Screenshots or screen recordings

Before	After

How to set up and validate locally

MR acceptance checklist

Evaluate this MR against the MR acceptance checklist. It helps you analyze changes to reduce risks in quality, performance, reliability, security, and maintainability.