I couldn't find an MSDS or a data sheet for the TivaGlo solutions but I did find a price - $771 for 5 gallons for all the various karats. That figures out to $40/liter. Therefore, you can't expect the gold content of a new solution to be more than about .8 grams/liter. It could be less. No matter what the karat is, they should contain the same amount of gold, when new. The lower karat colors are obtained by adding certain base metals, like copper, to the solution.
http://www.goldtouchinc.com/tankplating/TankPlatingChemicalPrices.html
Although there may be exceptions, most all of the non-cyanide gold solutions I've seen are made from sodium gold sulfite and operate at an alkaline pH - usually about 9.5. The gold complex is unstable at an acid pH and will decompose. When the pH is lowered to the acid side, the gold will drop out as a fairly pure brown sponge that is very easy to filter. As a sample, I would take about 100-200 ml of the gallon sample and add 10%, by volume, sulfuric acid to it, under a fume hood. When diluting the sulfuric, add the acid to the water. You can also use battery acid (about 35% sulfuric - buy at an auto parts store) diluted about 4 or 5 times with water. Add the sulfuric to the gold sample, in small increments, with a little stirring, and see what happens. If this is a normal sulfite solution, the gold will all drop out at some point. It shouldn't take much acid. Once you reach a certain pH, brown powder should start forming in the solution. Since there is no simple way to test the solution for gold, it's hard to estimate how much acid you'll need. Also, the solution may be buffered in order to stabilize it at the operating pH. It may take a little extra acid to overpower the buffer. I don't think a little extra acid will hurt.
I have read that there are some newer gold sulfite solutions that can be operated at an acidic pH. If so, it is possible that the sulfuric method given above may not work. If this is the case, I would lower the pH of a small 200 ml sample to about 2, or so, with weak sulfuric (maybe 5 or 10 ml of 10% sulfuric acid to your 200 ml sample) and add some 325 mesh zinc powder. The 325 mesh zinc powder is available from Ebay in 1# units and is inexpensive. Buy only 325 mesh. Add the zinc in small pinches, with stirring, since it can react violently, especially when you add too much at one time or, if you have started with too much acid. Wear gloves and a face shield. The zinc should precipitate all of the gold. Use a slight excess of zinc. You can tell this when the zinc no longer reacts with the acid and the precipitate appears as a gray powder. Filter and rinse. Transfer the powder to a beaker, cover with water, and add a little full strength battery acid. The acid will dissolve the gray zinc and you will be left with brown gold powder. It helps to heat it slightly at the end.
If you try to plate the gold out, I doubt if you'll be able to get it all. These "color" gold baths have very little gold to start with and most are only designed to plate for short periods of time. As the gold content gets lower, the plating efficiency gets lower, and the current splits more and more water at the cathode. At some point, all the current will be used in water-splitting and almost none will be used to deposit gold.
If you evaporate the solution, you will end up with all the chemicals in a dried out condition. In order to recover the gold, you will have to redissolve them and will probably then have to use one of the methods above. In other words, you will be right back to where you started from.
Try the sulfuric method first. It is by far the simplest method and I would expect it to work. Use small samples in your experiments. You don't want to screw up the entire solution, in case the method doesn't work. Let me know what happens.
