Euler_Explicit with adaptable step, how can i graph my solution?

I have the following system of differential equations (it is a simplified model for the activation of a neuron, but it is not really important for the question):

dV/dt = V - V^3/3 - n^2 + Iapp where Iapp is a function of time or a constant. And:

dn/dt = 0.1(2/(1-exp(-5V))-n) I want to solve this system using Euler's explicit method that I have to make myself and, also, using ode45.

I have done as follows:

 function x = nagumo1(t, y, f)

 Iapp = f(t);

 e = 0.1;

 F = 2/(1+exp(-5*y(1)));

 n0 = 0;

 x = zeros(1, 2);
 x(1) = y(1) - (y(1).^3)/3 - y(2).^2 + Iapp;
 x(2) = e.*(F + n0 - y(2));
 end

which calculates the derivatives at time t, with initial conditions y = [V, n]

Now my function Euler Explicit is:

 function x =  EulerExplicit1(V0, n0, tspan, Iapp)

 format shorteng;

 erreura = 10^-2;
 erreurr = 10^-2;

 h = 0.1;                             

 to =tspan(1, 1) + h;                 
 temps = to;
 tf = tspan(1, 2);

 y = zeros(1,2);
 yt1 = zeros(1, 2);
 yt2 = zeros(1, 2);
 y = [V0, n0];

 z = y;

 i = 1;

 s = zeros(1, 2);
 st1 = zeros(1, 2);

 for t = to:h:(tf - h)

      s = nagumo1(to+i*h, y, Iapp);
      y = y + h*s;
      yt1 = y + (h/2)*s;
      st1 = nagumo1(to+(i*h+h/2), yt1, Iapp);
      yt2 = yt1 + (h/2)*st1;

     if abs(yt2-y)>(erreura+erreurr*abs(y))
         test = 0;
     else
          h = h*2;
          test = 0;
     end

     while test == 0

          if abs(yt2-y)>(erreura+erreurr*abs(y)) & h>0.005
              h = h/2;
              s = nagumo1(to+i*h, y, Iapp);
              y = y + h*s;
              yt1 = y + (h/2)*s;
              st1 = nagumo1(to+i*h+h/2, yt1, Iapp);
              yt2 = yt1 + (h/2)*st1;
          else
              test = 1;
          end
     end

     temps(i)=to+i*h;
     z = [ z ; y ];
     i = i+1;

 end

 x = zeros(size(z));
 x = z;

 disp('Number of iterations:');
 disp(i);

 temps(i) = tf;
 plot(temps, x(:, 1:end), 'r');
 grid;

end

My main problem is the plotting of the solution. I have difficulty creating the time vector since the step is varying. I would also like to see how I could implement ode45 for this.