Base manifest doesn't allow benchmark groups except "default" benchmark group

Where applicable, please fill out the details below to help us troubleshoot the
issue that you are facing. 🍰

How to reproduce the issue
Using script:

python3 -m venv testvenv
source testvenv/bin/activate
pip install pyperformance
testvenv/bin/pyperformance run -b math

What you expected to happen
The tool will run all the math benchmarks producing output like:

[1/3] float...
.....................
float: Mean +- std dev: 48.3 ms +- 1.6 ms
[2/3] nbody...
.....................
nbody: Mean +- std dev: 46.6 ms +- 1.7 ms
[3/3] pidigits...

What actually happens

Collecting pyperformance
  Using cached pyperformance-1.11.0-py3-none-any.whl.metadata (2.0 kB)
Collecting pyperf (from pyperformance)
  Using cached pyperf-2.7.0-py3-none-any.whl.metadata (6.0 kB)
Collecting packaging (from pyperformance)
  Using cached packaging-24.1-py3-none-any.whl.metadata (3.2 kB)
Collecting psutil>=5.9.0 (from pyperf->pyperformance)
  Using cached psutil-6.0.0-cp36-abi3-manylinux_2_12_x86_64.manylinux2010_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (21 kB)
Using cached pyperformance-1.11.0-py3-none-any.whl (4.2 MB)
Using cached packaging-24.1-py3-none-any.whl (53 kB)
Using cached pyperf-2.7.0-py3-none-any.whl (139 kB)
Using cached psutil-6.0.0-cp36-abi3-manylinux_2_12_x86_64.manylinux2010_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl (290 kB)
Installing collected packages: psutil, packaging, pyperf, pyperformance
Successfully installed packaging-24.1 psutil-6.0.0 pyperf-2.7.0 pyperformance-1.11.0
WARNING:root:no benchmark named 'math'
Python benchmark suite 1.11.0



ERROR: No benchmark was run

Commencez par reproduire le problème avec python3 -m venv testvenv, installer pyperformance et exécuter testvenv/bin/pyperformance run -b math. Suivez le point d’entrée de pyperformance run et la gestion du base-manifest pour trouver pourquoi seul le groupe de benchmarks par défaut est accepté ; la tâche est terminée lorsque le groupe math exécute ses trois benchmarks au lieu d’indiquer qu’aucun benchmark n’a été exécuté.