The concept for these sorts of spacers is to find some suitable cylindrical insulator and figure out a way to attach some metal contacts on each end of the insulator.
I used hollow plastic tube. As an expedient, you can use a Papermate ball point pen body. The metal contacts are epoxied into the ends of the tube. The contacts themselves consist of a brass screw, brass nut, and AMP/Tyco crimp-on ring terminal. You can either crimp or solder your wire to the ring terminal. Slide the ring terminal onto the brass screw. Then thread on the brass nut to hold the ring terminal in place. Epoxy the end of the brass screw in the hollow plastic tube.
I actually cut my plastic tube slightly overlength (by about 1mm-2mm). That allowed me to file a flat on each brass screw head to increase the contact area. After filing, the overall length of the spacer was exactly AA length.
Kramer's approach also looks simple, using readily available materials. You can get brass rod from onlinemetals.com. In fact, you could even dispense with the brass rod. You can use some brass wood screws at each end of the wood spacer.
If you go with a hobby charger and decide to set up a parallel charging cradle arrangement, here is an example using Bulgin holders:
I used Deans 12 gauge noodle wire to connect the Bulgin holders in parallel. The noodle wires terminate in 15A banana plugs, which plug into the output jacks on your typical hobby charger.
The main issue with using a relatively low powered hobby charger is that you may run into difficulty charging at 1C if you want to charge many cells at the same time (e.g., 5xIMR26500). A 1C charge rate for an IMR26500 is 2.3A. Five IMR26500s in parallel means that your hobby charger needs to deliver 5*2.3A = 11.5A. If your charge rate can't hack it, then you kind of defeat the purpose of the whole arrangement -- rapid recharging of multiple Li-ions. If the hobby charger's max charge current is only 5A, for example, then you'd charge 5xIMR26500 at about 0.4C, not 1C.
If you plan to charge only a couple of cells in parallel, then most likely any hobby charger has enough "oomph".
You also typically need to supply an external power supply to run the hobby charger. Think of the hobby charger in the same way as a flashlight DC-DC converter. The hobby charger has a certain "driver efficiency".
For example, you might have to feed the charger 20V and 3A so that the charger can deliver 11.5A and up to 4.20V to 5xIMR26500. Bottom line is that your power supply has to be able to deliver more power than what the charger sends to the cells because the charger isn't 100% efficient. In the notional example above, you need a 60W power supply to charge 5xIMR26500, even though the max power that gets sent to the IMRs is only 48W.