1.)
Yes, there is a stereo effect if you pan the mic one way and the D.I the other. But it’s not a great way of doing stereo. Why? Because it’s the same performance; the two sides will be mostly phase coherent. The differences between the tracks will be panned (ie the ways in which the D.I. tone differ from the amp tone - maybe one side will have harmonics the other doesn’t due to distortion), but the similarities between the tracks (and there will be many) will be drawn across the stereo image between the pan positions of the two signals, in much the same way as, say, panned drum overheads give you the common sounds in both channels (kick, snare etc) in the middle even when hard panned.
So in practice what you’ll get is a mostly mono sound with certain frequencies leaning one way or the other.
And that’s if the SM57 and D.I. signals are in time; what’s more likely is that because of the speed of sound between the speaker and SM57, there will be a fixed time offset between the two signals. This will make it more stereo and less mono, but will sound worse because now you’ll also have a weird, comb-filtery tone - and your ears, which are really clever, will know they’re hearing the same performance twice.
2.)
It goes back to phase coherent vs phase incoherent. If one performance is split into two signal chains and recorded to two tracks, it’s still only one performance; the same timing, dynamics, even down to how each string moves and that’s why you get the effect described in my answer to question one - it produces two phase-coherent signals.
When you record two separate performances, even if you use the exact same amp settings, mic position etc, you’ll end up with two phase incoherent signals because on a millisecond by millisecond bases the signals will be different - because of the way you play the guitar and because between the two takes from moment to moment each guitar string will be in a different position in its oscillation if you could compare the two performances side by side.
This is why there’s no real substitute for proper doubling of parts - it’s the only way to minimise phase coherency between the tracks.
3.)
Chorus can produce stereo because it works by constantly changing the pitch and time differences between the dry signal and the wet signal. So at any given moment, there is a fixed relationship between the dry signal and the wet signal that, if left unchanged, your ears would hear as a strange, hollow sound - comb filtering. But because the differences are always shifting, your ear doesn’t lock onto that strange sound and instead you hear a shimmery, chorus effect. So if you have to create stereo from mono, this is a good way to do it that minimises downsides. (as long as you WANT a chorus sound and you pick a chorus setting that doesn’t sound dreadful when it collapses to mono).