Fix zero spin conversion

This fixes the issue in #588 (closed). It just needed some rearranging of the order in which the various parameters are added and calculation are performed when filling in the missing pieces of the result object.

bilby/gw/conversion.py

@@ -1154,7 +1154,6 @@ def compute_snrs(sample, likelihood):
             matched_filter_snrs = {
                 ifo.name: [] for ifo in likelihood.interferometers}
             optimal_snrs = {ifo.name: [] for ifo in likelihood.interferometers}
-            pdb.set_trace()
             for ii in tqdm(range(len(sample)), file=sys.stdout):
                 signal_polarizations =\
                     likelihood.waveform_generator.frequency_domain_strain(