run! does not allow for heterogeneous primitive types which are subtypes or a homogeneous abstract type

[fieldofnodes]

Describe the bug

Mini scenario

Let adata = [:abstract_type_var] where :abstract-type-var will have an abstract type SuperType and the outcomes of :abstract-type-var will have types such as A <: SuperType or B <: SuperType
One would expect after running the run! command to collect data that the column would look like:
adf.abstract_type_var = [A(),B(),C()] where each row has a different primitive type, but they are all the same abstract type.

Statement

Function: run! does not allow for heterogeneous primitive types which are subtypes or a homogeneous abstract type.

Minimal Working Example

abstract type SuperType end
struct A <: SuperType end
struct B <: SuperType end

@agent struct TestAgent(NoSpaceAgent)
    type_agent::SuperType
end


        
# Set up model
function initialize_model(;num_agents = 10, seed = 123)

    rng = MersenneTwister(seed)
    properties = ()
    
    model = StandardABM(TestAgent;
        agent_step! = agent_step!,
        properties = properties, 
        rng = rng, 
        scheduler = Schedulers.Randomly(), 
        warn = false
    )


    # Add agents
    for _ in Base.OneTo(num_agents)
        add_agent!(
            TestAgent, 
            model, 
            rand(abmrng(model),[A(),B()]))
    end

    return model

end

function agent_step!(agent,model)
    agent.type_agent = rand(abmrng(model),[A(),B()])
end


    
# Run a sample
stable_params = (;num_agents = 10,seed = 2314)
model = initialize_model(;stable_params...)
adata = [:type_agent]
steps = 60
adf, _ = run!(model, steps; adata)

This gives the following output

julia> adf, _ = run!(model, steps; adata)
┌ Error: Error adding value to column :type_agent. Maybe you forgot passing `promote=true`?
└ @ DataFrames C:\Users\jonathan.miller\.julia\packages\DataFrames\kcA9R\src\dataframe\insertion.jl:369
ERROR: MethodError: Cannot `convert` an object of type B to an object of type A
The function `convert` exists, but no method is defined for this combination of argument types.

Closest candidates are:
  convert(::Type{T}, ::T) where T
   @ Base Base.jl:126

Stacktrace:
  [1] push!(a::Vector{A}, item::B)
    @ Base .\array.jl:1244
  [2] _append!(a::Vector{A}, ::Base.HasShape{1}, iter::Vector{SuperType})
    @ Base .\array.jl:1322
  [3] append!(a::Vector{A}, iter::Vector{SuperType})
    @ Base .\array.jl:1313
  [4] _append_or_prepend!(df1::DataFrame, df2::DataFrame; cols::Symbol, promote::Bool, atend::Bool)
    @ DataFrames C:\Users\jonathan.miller\.julia\packages\DataFrames\kcA9R\src\dataframe\insertion.jl:308
  [5] _append_or_prepend!
    @ C:\Users\jonathan.miller\.julia\packages\DataFrames\kcA9R\src\dataframe\insertion.jl:239 [inlined]
  [6] append!(df1::DataFrame, df2::DataFrame)
    @ DataFrames C:\Users\jonathan.miller\.julia\packages\DataFrames\kcA9R\src\dataframe\insertion.jl:93
  [7] collect_agent_data!(df::DataFrame, model::StandardABM{…}, properties::Vector{…}, step::Int64; kwargs::@Kwargs{…})
    @ Agents C:\Users\jonathan.miller\.julia\packages\Agents\eKASH\src\simulations\collect.jl:474
  [8] collect_agent_data!
    @ C:\Users\jonathan.miller\.julia\packages\Agents\eKASH\src\simulations\collect.jl:458 [inlined]
  [9] run!(model::StandardABM{…}, n::Int64; when::Int64, when_model::Int64, mdata::Nothing, adata::Vector{…}, obtainer::Function, showprogress::Bool, init::Bool, dt::Float64)
    @ Agents C:\Users\jonathan.miller\.julia\packages\Agents\eKASH\src\simulations\collect.jl:137
 [10] top-level scope
    @ c:\Users\jonathan.miller\Projects\genetic_modelling\src\julia-version\insect_sex_mating\src\testing_insect_sex_mating.jl:308
Some type information was truncated. Use `show(err)` to see complete types.

This is possible directly in DataFrames.jl

using DataFrames
vec_types = [A(),B()]
df = DataFrame(abstract_type_var = vec_types)
df

gives

2×1 DataFrame
 Row │ abstract_type_var 
     │ SuperType
─────┼───────────────────
   1 │ A()
   2 │ B()

Agents.jl version

Agents v6.1.9

Julia version

Version 1.11.0 (2024-10-07)