Not sure what you mean about soft patching - I know how to do it, but how would it help in this situation? The MON and FOH would still both be trying to use the same faders, right?
The LS-9 has 4 layers. The first 2 layers are input layers, the third layer is outputs and the fourth layer is custom faders. A LS-9 32 is having 2 layers of inputs which gives you a total of 64 channels. But only the first 32 channels(1 to 32) on the first layer are having pres. Unless you use the 168 ES for the 2nd layer of inputs which are 33 to 64, then you'll have 64 pres for all 64 channels.
Despite the second input layer has no pres, it doesn't mean they are useless. In fact, the second layer of inputs are very useful in a situation which you do monitors from FOH, and specially your case. Go to the second input layer and patch the input source corresponding to the first layer, example: patch input source of ch 33 to ch 1, ch 34 to ch 2, ch 35 to ch 3 and so on. Now you have the second layer exactly the same copy of the first layer, an identical 32 inputs of 2 layers, this is what soft patching about. You are only sharing the pres of the first layer with the second layer, all others are operate separately. EQ, comps, gates, aux sends, pan, faders, which you can control separately directly from the console or studio manager or the ipad.