The one you left out was both! I've been studying Emily Remler and she uses pick for the fast improv then moves it to her mouth and uses fingers for chord melody. I've seen Pass do likewise but I guess this requires a convenient pause to make the change.
I use my fingers mainly as well; I was a classical guitarist originally as well and now prefer chord melody. Joe Pass studied Carcassi's etudes early on in his life and probably influenced his style greatly.
Matteo Mancuso dispels any doubt that fingerstye is in anyway limtied speed-wise. Granted, only very few can likely attain his unusual technique, but in terms of sheer notes per BPM he has broken through msot upper limits already. Having said that, I think it is fair to say that using a pick lends itself more naturally to rhythmic urgency and vitality. Again, not necessarly always - someone will always come along who can overcome the normative liitations of any approach. But think of the subtle difference in 8th note swing feel between say, Lenny braeu and Pat Martino. Lenny is an undisputed genius, but most listeners would notice Pat's rhythmic mojo as being somewhat more native to using a plectrum.
The guitar cord can definitely make a difference in tone if the cable capacitance and guitar / amp impedances roll off the high frequencies. This happens more with longer leads and higher impedances. Some cables are lower capacitance and so have less of an impact. None of this matters if you play with a very mellow sound, but the brighter your sound, the more you're likely to notice. A guitarist who uses heavy distortion or fuzz is also less likely to notice a high frequency roll off, because distortion adds harmonics after the guitar cord. Is high frequency roll off a bad thing? No, it's just a difference. Brian May famously plays with a "curly cord" which adds a series inductance to the cable's parallel capacitance and multiplies the effect, because it gives him the sound he wants.
