fix: Avoid ZeroDivisionError if output_tokens is 0

[ccamacho]

This commit validates that when calculating the results if output_tokens is 0 then both itl and tpot must be None.

[ccamacho]

When testing HF TGIS im hitting

Traceback (most recent call last):
  File "/usr/lib64/python3.9/multiprocessing/process.py", line 315, in _bootstrap
    self.run()
  File "/usr/lib64/python3.9/multiprocessing/process.py", line 108, in run
    self._target(*self._args, **self._kwargs)
  File "/var/mnt/data/sw/llm-load-test/user.py", line 83, in run_user_process
    result = self.make_request(test_end_time)
  File "/var/mnt/data/sw/llm-load-test/user.py", line 50, in make_request
    result = self.plugin.request_func(query, self.user_id, test_end_time)
  File "/var/mnt/data/sw/llm-load-test/plugins/hf_tgi_plugin.py", line 117, in streaming_request_http
    result.calculate_results()
  File "/var/mnt/data/sw/llm-load-test/result.py", line 49, in calculate_results
    self.itl = (1000 * (self.end_time - self.first_token_time)) / (
ZeroDivisionError: float division by zero
2024-08-18 21:48:10,576 INFO     root MainProcess Adding 1 entries to dataset queue

[kpouget]

this looks quite like

Suggested change

	if self.output_tokens is not None and self.output_tokens > 0:
	if self.output_tokens:

as output_tokens cannot be negative.
Not sure what's better...

(same below)

[ccamacho]

ouuhhh I got the actual error self.output_tokens is 1 and self.output_token exists