Broken ffmpeg commands for Intel GPUs on Linux

### Classifications

- [x] Running `python ./jellybench_py/hwi.py` manually (within the Poetry shell) works and produces correct results.
- [ ] This happens on a Virtual Hardware (VM, WSL, ...)
- [x] The issue occurs during the benchmark section of the script (after `Starting Benchmark...`).
- [x] A server issue or faulty server response is causing the problem.
- [ ] jellybench was run in `--debug` mode (recommended for issue reports).
- [x] I am using the latest version (branch `develop`) of jellybench. I understand older versions are not supported.

### Description

When run jellybench with `--nocpu` option the following exception is thrown when starting benchmark

### Error Code or Output

```shell
We will do 72 tests.
Continue (y/n): y

Status: Testing Workers: 01 Last Speed: 00.00 0% [ ] ETA: --:--:--
Traceback (most recent call last):
File "/home/user/.cache/pypoetry/virtualenvs/jellybench-py-ZcGqwGZJ-py3.12/bin/jellybench", line 6, in
sys.exit(main())
^^^^^^
File "/home/user/jellybench_py/jellybench_py/core.py", line 760, in main
return cli()
^^^^^
File "/home/user/jellybench_py/jellybench_py/core.py", line 727, in cli
valid, runs, result = benchmark(
^^^^^^^^^^
File "/home/user/jellybench_py/jellybench_py/core.py", line 186, in benchmark
output = worker.workMan(total_workers, ffmpeg_cmd)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/user/jellybench_py/jellybench_py/worker.py", line 146, in workMan
"rss": workrss,
^^^^^^^
UnboundLocalError: cannot access local variable 'workrss' where it is not associated with a value
```

### Steps to Reproduce

1. start jellybench with `--nogpu` flag
2. start benchmark
3. see error

### Operating System

Linux

### System Architecture

x86_64 (64-bit)

### Python Version

3.12.3

### CPU[s]

Intel Core i5-10400F

### GPU[s]

Intel Arc A750

### Additional Context

Benchmark starts properly without the `--nocpu` flag

### Possible Solution

_No response_

Start in jellybench_py/worker.py around line 146, then trace the benchmark call from jellybench_py/core.py around lines 186 and 727. Reproduce the reported benchmark run with the --nocpu or --nogpu option and confirm the UnboundLocalError no longer occurs and the benchmark completes.

Written by the indexing model from the issue text.