examples:
Discrete action space
python reinforce_main_is.py --task Isaac-Cartpole-Showcase-Box-Discrete-Direct-v0 --algorithm reinforce --num_envs 64 --n-episodes 500 --headless --device cuda:0
continious action space
python reinforce_main_is.py --task Isaac-Cartpole-v0 --algorithm reinforce --num_envs 64 --n-episodes 500 --headless --device cuda:0