This PR implements concurrency control on top of the resource scheduler and introduces a new type of RESOURCE and a new WORKLOAD setting to limit the number of concurrent threads for a specific workload.

Usage example:

CREATE RESOURCE cpu (MASTER THREAD, WORKER THREAD)
CREATE WORKLOAD all
CREATE WORKLOAD admin IN all SETTINGS max_concurrent_threads = 10
CREATE WORKLOAD production IN all SETTINGS max_concurrent_threads = 100
CREATE WORKLOAD analytics IN production SETTINGS max_concurrent_threads = 60, weight = 9
CREATE WORKLOAD ingestion IN production

This configuration example provides independent CPU slot pools for admin and production. The production pool is shared between analytics and ingestion. Furthermore, if the production pool is overloaded, 9 of 10 released slots will be rescheduled to analytical queries if necessary. The ingestion queries would only receive 1 of 10 slots during overload periods. This might improve the latency of user-facing queries. Analytics has its own limit of 60 concurrent thread, always leaving at least 40 threads to support ingestion. When there is no overload, ingestion could use all 100 threads.

WARNING: Slot scheduling provides a way to control query concurrency but does not guarantee fair CPU time allocation yet. This requires further development of CPU slot preemption and will be supported in the follow-up PR.

Changelog category (leave one):

• New Feature

Changelog entry (a user-readable short description of the changes that goes to CHANGELOG.md):

Added CPU slot scheduling for workloads, see https://clickhouse.com/docs/operations/workload-scheduling#cpu_scheduling for details

Documentation entry for user-facing changes

• Documentation is written (mandatory for new features)