Instrument broadening from Spectrum1D: fix weighting

There seem to be two y-axis conventions at work; for "sparse" data
that has not been binned, the weights indicate a symmetry weighting or
possible neutron scattering intensity. For data in a spectrum, there
is also a weight for the bin width to compensate for bin-size effects.

The variable-width broadening machinery currently expects the
former (which makes sense for adaptive DOS broadening) and converts to
the latter (which is the expected form for spectrum
objects). Broadening _from_ a spectrum object therefore needs to
account for this.

euphonic/spectra.py

@@ -583,9 +583,15 @@ def broaden_with_polynomial(self: T,
         else:
             raise ValueError('width_convention must be "std" or "fwhm".')
 
+        bin_widths = np.diff(self.get_bin_edges())
+        if not np.all(np.isclose(bin_widths, bin_widths[0])):
+            raise ValueError('Not all bins are the same width: this method '
+                             'requires a regular sampling grid.')
+
         y_broadened = polynomial_broadening(
             self.get_bin_edges(), self.get_bin_centres(), polynomial,
-            self.y_data.magnitude,
+            (self.y_data.magnitude
+             * bin_widths[0].to(1/self.y_data.units).magnitude),
             width_unit=width_unit, width_lower_limit=width_lower_limit,
             adaptive_error=adaptive_error)