Too restrictive typing (Float64)

[tjungni]

Many functions currently expect arguments to be Float64 instead of Real.
Constants will ideally be defined as irrationals so they don't force type promotions in calculations with other values.
For example,

function OutgoingFermionSpinor(
    mom::T, mass::Float64
) where {T<:AbstractLorentzVector{TE}} where {TE<:Real}
    return OutgoingFermionSpinor(_build_particle_booster(mom, mass))
end

expects the mass to be Float64 but I defined the electron mass as an irrational constant in my code and tried to pass that as mass.

[szabo137]

similar issue for element types: #60