Prevent allocs for FD{Int128} by freeze max_exp10

Before this commit, reinterpret(FD{Int128,f}, x) allocates because:
 1. max_exp10(Int128) widens the Int128 to a BigInt, which causes
    allocations.
 2. The result of max_exp10(Int128) isn't getting const-folded away, so
    the above widen operation occurs every time it's called at runtime.

This commit simply "freezes" the result for Int128s via a top-level
`@eval` statement.

Author: NHDaly

src/FixedPointDecimals.jl

@@ -484,6 +484,7 @@ function max_exp10(::Type{T}) where {T <: Integer}
 end
 
 max_exp10(::Type{BigInt}) = -1
+@eval max_exp10(::Type{Int128}) = $(max_exp10(Int128))  # Freeze this, since it's not getting Const folded.
 
 """
     coefficient(::Type{FD{T, f}}) -> T