Run tpcc/ycsb benchmarks with enabled shared threads

**branche**: main
**benchmarks**: tpcc/ycsb
**configs**:
```
actor_system_config:
    ... # autoconfig fields
    use_shared_threads: true
```

It is necessary to update the configuration wherever auto-config is used.
Of particular interest are results obtained with a small number of cores (4) on the compute node and with the regular number of cores