LocalCluster does not respect n_workers and threads_per_worker

[jackmead515]

Describe the issue:

When running:

dask_client = LocalCluster(n_workers=2, threads_per_worker=2)

It still spins up as many as my cpu_count.

Needs to respect explicit settings given by the user.

Minimal Complete Verifiable Example:

# Put your MCVE code here

Anything else we need to know?:

Environment:

• Dask version: 2025.5.1
• Python version: 3.10
• Operating System:
• Install method (conda, pip, source): pip