Bound Julia version due to RuntimeGeneratedFunction problem on 1.6

```julia
using RuntimeGeneratedFunctions
RuntimeGeneratedFunctions.init(@__MODULE__)
ex = :(function foo(a, b)
    goo = function (c, d)
    end
end)
fun = @RuntimeGeneratedFunction(ex)
fun(1, 2)
```
gives
```julia
julia> fun(1, 2)
ERROR: The function body AST defined by this @generated function is not pure. This likely means it contains a closure or comprehension.
Stacktrace:
 [1] (::RuntimeGeneratedFunction{(:a, :b), var"#_RGF_ModTag", var"#_RGF_ModTag", (0x1c41534f, 0x444455e2, 0x6114b514, 0xa7b18630, 0x10e959a2)})(::Int64, ::Int64)
   @ RuntimeGeneratedFunctions ~/.julia/packages/RuntimeGeneratedFunctions/6v5Gn/src/RuntimeGeneratedFunctions.jl:124
 [2] top-level scope
   @ REPL[47]:1
```

Author: YingboMa

src/systems/clock_inference.jl

@@ -163,6 +163,9 @@ end
 function generate_discrete_affect(syss, inputs, continuous_id, id_to_clock;
                                   checkbounds = true,
                                   eval_module = @__MODULE__, eval_expression = true)
+    @static if VERSION < v"1.7"
+        error("The `generate_discrete_affect` function requires at least Julia 1.7")
+    end
     out = Sym{Any}(:out)
     appended_parameters = parameters(syss[continuous_id])
     param_to_idx = Dict{Any, Int}(reverse(en) for en in enumerate(appended_parameters))

test/clock.jl

@@ -112,10 +112,12 @@ eqs = [yd ~ Sample(t, dt)(y)
        =#
        ]
 @named sys = ODESystem(eqs)
-ss = structural_simplify(sys)
-prob = ODEProblem(ss, [x => 0.0, y => 0.0], (0.0, 1.0),
-                  [kp => 1.0; z => 0.0; z(k + 1) => 0.0])
-sol = solve(prob, Tsit5(), kwargshandle = KeywordArgSilent)
+ss = structural_simplify(sys);
+if VERSION >= v"1.7"
+    prob = ODEProblem(ss, [x => 0.0, y => 0.0], (0.0, 1.0),
+                      [kp => 1.0; z => 0.0; z(k + 1) => 0.0])
+    sol = solve(prob, Tsit5(), kwargshandle = KeywordArgSilent)
+end
 # For all inputs in parameters, just initialize them to 0.0, and then set them
 # in the callback.