Ensure templated floating-point values are used throughout

[mlxd]

Currently, Lightning has support throughout to enable strict use of templates for floating point values (float or double), and supports one type from Python through to CPP.

For the following function, we explicitly enable this as double, rather than the expected class template fp_t.

pennylane-lightning/pennylane_lightning/src/simulator/Measures.hpp

Line 330 in e9fa573

std::vector<double> bucket(N);

The goal is to ensure this is updated to use the fp_t value instead of explicitly using doubles or floats in the Measures class.