Bug: Fixing worker-state initialization bug

[justinwlin]

Description

Running runpod --help crashes the runpod python sdk b/c we are starting a multithreaded process is causing the crash.

The fix is to not create the object until methods are specifically called. So at initialization we should initialize to None and only when the classes are acted on, do we initialize and member variables for multithreading

Debug Stack:

Error when I ran --help

(venv) justinwlin@Justins-MBP-2 runpod-python % runpod
...
RuntimeError: 
        An attempt has been made to start a new process before the
        current process has finished its bootstrapping phase.

        This probably means that you are not using fork to start your
        child processes and you have forgotten to use the proper idiom
        in the main module:

            if __name__ == '__main__':
                freeze_support()
                ...

        The "freeze_support()" line can be omitted if the program
        is not going to be frozen to produce an executable.

        To fix this issue, refer to the "Safe importing of main module"
        section in https://docs.python.org/3/library/multiprocessing.html
       ...
    buf = self._recv(4)
  File "/Library/Frameworks/Python.framework/Versions/3.13/lib/python3.13/multiprocessing/connection.py", line 399, in _recv
    raise EOFError
EOFError
(venv) justinwlin@Justins-MBP-2 runpod-python % 

The main thing to look at is:

RuntimeError: 
        An attempt has been made to start a new process before the
        current process has finished its bootstrapping phase.

Looking deeper into why the reasoning is:

runpod --help
  ↓
imports runpod.cli
  ↓  
imports runpod.cli.groups.pod
  ↓
imports runpod.serverless.modules.rp_job 
  ↓
job_progress = JobsProgress()  # This line runs at import time
  ↓
JobsProgress.__init__() runs
  ↓
Manager() gets created  # ❌ Multiprocessing object created during import

rp_job.py

JOB_GET_URL = str(os.environ.get("RUNPOD_WEBHOOK_GET_JOB")).replace("$ID", WORKER_ID)

log = RunPodLogger()
job_progress = JobsProgress()

How to Test:

I wasn't sure how to test, so I just ran the tests again and had tests generated by claude for --help command for sanity test in the future.

Passing status quo cases

Showing that without the fix we can capture --help errors in the future:

[deanq]

After many iterations, I have decided to start from scratch to fix this without hacks. Thanks for the initial sweep, @justinwlin . Please refer to this PR #430 for the final fix.