The main point about it is the retention of context for the given sub-task. You have a main task and a sub-task, sometimes an interaction with both is necessary in parallel. The bottom sheet presents the sub-task while the main context is still visible, so as to aid the user's mental flow.
Here are a two examples.
The best example is Google Maps (or anything similar). Imagine having to switch between screens constantly. The advantage here is simultaneous interaction with two screens.
In this case it helps the user remember where he is about to upload his new file. Maybe he made a mistake and chose the wrong folder, this way he can still spot it while in the process.
Your example of Todoist is definitely debatable, as it seems cramped and too small. We can only guess why they used it, maybe wanted to go with "cool" new tech? Who knows.
But the point is that the bottom sheet definitely has its UX advantages for specific kinds of tasks.
Edit: In response to the question in the comments
Does it make sense for my case?
As long as it really offers an advantage (e.g. in terms of context or error prevention), it would be a good idea. Though, one should be careful to not introduce unexpected disadvantages with this functionality; meaning to think through the standard user flow and see where it helps or hinders the user (weight advantages vs disadvantages).
For example the screenshot of Todoist makes it look very cramped and too small, where I would assume the advantage of context is somewhat damaged by the (presumably) bad UX.