Fix random seed in T1 tests

[yaelbh]

Identical to #69, but for T1 tests.
Are randomized benchmarking tests deterministic? Considering that the experiment itself involves randomizations. @ItamarGoldman @gadial @ShellyGarion

[ShellyGarion]

The tests for RB @gadial @ItamarGoldman (as well as for QV @dekelmeirom) are divided into two parts:

1. Test of the circuit generation - checks that the random circuits satisfies the conditions (the metadata is correct, in RB the circuit is equivalent to identity etc.). Here there is a fixed seed, see e.g.:
   https://github.com/Qiskit/qiskit-experiments/blob/d8cb33e1b8652458652a40011d5578179d186233/test/test_rb.py#L48

2. Test of the fitters [WIP] - this is based on pre-generated data which is stored in json files. The code for generating the json files will also be provided as part of the test directory, and should include a fixed seed.