Addressing one by one ( I don't mean to impose my views on you, I am just trying to present some points) :
1) If the same loco has to be used after reversal, then it involves the following steps ( excluding the final attachment of the loco to the train):
a) After loco is detached, the new crew has to take over the locomotive and perform the usual diagnostics (brake power etc) before starting up the locomotive. Usually takes a few minutes.
b)...
more... Controllers have to wait for line clearance and set the points, and block the respective lines. This process is usually the time consuming one, and at busy junctions, can cause a lot of delays.
c) Once the loco starts the process of switching ends, the crew has to switch cabs and possibly swap pantographs too twice, once for each reversal. This again eats up a few minutes.
d) The loco has to travel from one end of the station to the other.
If a fresh loco is used, step a is already done by the time the train even arrives. step 2 is no longer necessary and at busy junctions, this means a lot as this step is the most time consuming step. And step 'c' is also not necessary as the loco doesn't have to reverse multiple times. This saves quite a lot of time. At busy junctions like BZA, a new loco is attached within a few minutes for many trains that are having reversals, even at peak hours.
2) Earlier, when E-locos were not so abundant, pretty much every train having a reversal used the same loco after reversal. But as the number of locos started increasing and the availability increased, more and more trains started getting loco change when reversal is required. As KGP trip shed has an abundance of SRC WAP's now, they are able to provide a new loco here and they are preferring that, as it's convenient and faster.