Feature request: Add seed support to stim.Tableau.random() for reproducibility #974
Description
I'm currently using stim.Tableau.random(n_qubits) as part of a benchmarking/stochastic optimization routine, and I've noticed that the method doesn't support passing a seed for reproducibility.
Currently, calling stim.Tableau.random(n_qubits) gives a different tableau at every run, and there appears to be no way to fix the randomness deterministically.
Are there any alternative solutions by any chance? Otherwise, would it be possible to add this feature?