Just a couple of thoughts on the 2-tray or 2-developer system and other contrast-controlling techniques at the developing stage.
The 2-developer system is a tried and true way of achieving intermediate contrast grades. The developers can be mixed to make one tray of developer with an intermediate activity, or used one after the other.
Ansel Adams recommended using the softer developer first because he transferred the prints directly from one developer to the other without an intermediate water bath. The carry-over of Metol only to the more powerful MQ developer had no significant effect on the developer activity over a longer printing session. Also, the softer-working developer probably is less alkaline, again making the transfer from soft to hard developer insignificant whereas the reverse would soon "soup up" the soft developer by changing the pH and adding Hydroquinone to the mix.
With a water bath between, I can see no reason why either one first would make a difference. However, I have always used the soft developer first (Selectol Soft, or the like, diluted 1:1 or more) for 1-3 minutes and then the harder develper second. I also transfer directly from one developer to the other. I don't need to waste extra darkroom time with an intermediate water bath that way. I have found that 30 seconds is about the minimum time in the hard developer (with vigorous agitation) to maintain even development. (It might be interesting to do some tests: hard first, soft first for the same times, with and without the water bath and see if there is indeed a difference.)
As stated above, sometimes just the soft developer alone is the best solution. Again, in principle using the 2-developer method is just a way to alter the activity of the total development. There are also other methods.
I have for sometime also been "tweaking" my off-the-shelf developer with the addition of sodium carbonate and/or benzotriazole or potassium bromide to increase activity and restrain development in the whites, respectively. I simply make 10% solutions of each and add them "by guess or by golly" to the hard developer. Carbonate to increase activity (this speeds up the paper a bit) and benzotriazole or bromide to restrain the development in the whites, thus altering the paper scale (and slowing down the paper). A mix of the two gives deeper blacks and holds back the whites a bit too, thus increasing contrast without greying the whites.
A more concentrated developer will also give a marked increase in activity and, usually, more contrast (e.g. straight Dektol). The converse applies to diluting the soft-working developer. SS 1:3 with long developing times can really extend the paper scale.
Many people believe that increasing development time will increase contrast. However, with modern materials, this effect is minimal, being only significant before the paper reaches full development or with extremely long times (10+ minutes). Just take a look at the published curves for most papers. What increased development time does best is to effectively speed up the paper a bit. I use it to put the finishing touches on my basic exposure time. Sometimes 10 seconds makes a marked difference in the finished print.
All these techniques are fairly simple and, before the advent of VC papers, part of the arsenal of any good printer. They are still of most use to us few who prefer graded papers to VC. However, they are equally effective on VC papers, and prints made with split-filtration or lots of manipulations may even require them for optimum results.
I never use a intermediate water bath when using my 2 bath development for VC (or graded) enlarging paper. I cannot see any reason for doing so. As far as mixing the two baths together in one tray, I think that would reduce the ability to contropl contrast and basically would defeat the purpose of using 2 baths. If the point is to creat an "intermediate" developer, for whatever reason, ok, but if it's tonal control you're after, then I say keep the two baths seperate.