The answer is No, and the reason why it’s no gets to the very heart of Quantum Mechanics.
Take a look at the wiki article on Renninger’s Negative Result Experiment: https://en.m.wikipedia.org/wiki/Renninger_negative-result_experiment
Here’s what it demonstrates: in Quantum Mechanics, it’s the act of gaining information about a system that results in a loss of coherence between superpositions, not only the much more intuitive measurement events like smacking an electron with a photon. So it has nothing at all to do with how “gentle” you make your measurement.
Here’s the best example I can give:
In your double-slit setup, imagine if you only set up a detector in front of one slit. It’ll tell you whether a photon goes through that slit, but it obviously won’t interact with a photon going through the opposite slit in any way.
If your detector clicks, then you know your photon went through the left slit, and you’re unsurprised when all of the “click” runs of your experiment shows no interference.
But here’s the thing: If your detector doesn’t click, it still destroys the interference pattern. The absence of a click tells you with 100% certainty that the photon passed through the other slit, and so no interference can be recovered.
Information about the system is what matters.
This is a feature of QM that allows for things like the Elitzur–Vaidman bomb-tester to work despite appearing paradoxical. Negative-result or “interaction-free” measurements are just as valid as traditional measurements in reducing wavefunctions to observable results.