When using ROQs with bilby/bilby_pipe, if the prior is not suitable for the basis being used an uninformative error is sometimes returned rather than the more informative errors raised by the check [here](https://git.ligo.org/lscsoft/bilby/-/blob/master/bilby/gw/likelihood/roq.py#L576). I believe this happens when there is no overlap between the prior and ROQ basis, but I have not verified this. We should make sure the order of the different calls is correct and add a unit test to verify a helpful error is returned. **Example output** Example of the error <details> <summary> Traceback </summary> ``` Traceback (most recent call last): File "/local/condor/execute/dir_957284/condor_exec.exe", line 8, in <module> sys.exit(main()) File "/home/michael.williams/.conda/envs/nessai-review-gw/lib/python3.9/site-packages/bilby_pipe/data_analysis.py", line 379, in main analysis.run_sampler() File "/home/michael.williams/.conda/envs/nessai-review-gw/lib/python3.9/site-packages/bilby_pipe/data_analysis.py", line 264, in run_sampler self.result = bilby.run_sampler( File "/home/michael.williams/.conda/envs/nessai-review-gw/lib/python3.9/site-packages/bilby/core/sampler/__init__.py", line 190, in run_sampler sampler = sampler_class( File "/home/michael.williams/.conda/envs/nessai-review-gw/lib/python3.9/site-packages/bilby/core/sampler/base_sampler.py", line 252, in __init__ self._time_likelihood() File "/home/michael.williams/.conda/envs/nessai-review-gw/lib/python3.9/site-packages/bilby/core/sampler/base_sampler.py", line 441, in _time_likelihood self.log_likelihood(theta) File "/home/michael.williams/.conda/envs/nessai-review-gw/lib/python3.9/site-packages/bilby/core/sampler/base_sampler.py", line 841, in log_likelihood return Sampler.log_likelihood(self, theta) File "/home/michael.williams/.conda/envs/nessai-review-gw/lib/python3.9/site-packages/bilby/core/sampler/base_sampler.py", line 534, in log_likelihood return self.likelihood.log_likelihood() File "/home/michael.williams/.conda/envs/nessai-review-gw/lib/python3.9/site-packages/bilby/gw/likelihood/base.py", line 837, in log_likelihood return self.log_likelihood_ratio() + self.noise_log_likelihood() File "/home/michael.williams/.conda/envs/nessai-review-gw/lib/python3.9/site-packages/bilby/gw/likelihood/base.py", line 394, in log_likelihood_ratio self.waveform_generator.frequency_domain_strain(self.parameters) File "/home/michael.williams/.conda/envs/nessai-review-gw/lib/python3.9/site-packages/bilby/gw/likelihood/roq.py", line 390, in waveform_generator self._update_basis() File "/home/michael.williams/.conda/envs/nessai-review-gw/lib/python3.9/site-packages/bilby/gw/likelihood/roq.py", line 358, in _update_basis self._cache[basis_number_key] = np.arange(number_of_bases)[in_prior_range][0] IndexError: index 0 is out of bounds for axis 0 with size 0 ``` </details>