I disagree. An earth like object isn't going to bend light much, the maximum of 1G acceleration is just not that noticeable.
An earth mass blackhole is going to have 10, 1000s, or even millions of Gs depending on how close the light gets.
Seems like you'd get stars periodically blinking from places they shouldn't be, as light trajectory should have missed earth, but gets bent by the black hole. So while the black hole itself would be invisible at that distance, and occultations would be invisible, seeing stars in the wrong place would still be visible.
Yes, it will happen, but the radius of the effect is simply too small to see it.
You are only bending a minute faction of the light from the star traveling toward you. It's not enough light to see, and it's too small to resolve (unless you could resolve the black hole in the first place).
Think about it: The redirected light only has the "size" of the gravitation field in question. For an earth sized mass thats 9mm. Even if you did 9km (which wouldn't bend much) you couldn't see it, never mind 9mm.
From a distance there is no difference in lensing between a black hole and an regular object of the same mass.
From close there would be a difference, but if you could resolve something that small you would be able to directly image the object.