It's quite simple if you visualize the iterations:
1 # f'{1:1<1}', means start with 1, left align with 1 spaces filled with 1
22 # f'{2:2<2}', means start with 2, left align with 2 spaces filled with 2
333 # f'{3:3<3}', means start with 3, left align with 3 spaces filled with 3
4444 # f'{4:4<4}', means start with 4, left align with 4 spaces filled with 4
55555 # f'{5:5<5}', means start with 5, left align with 5 spaces filled with 5
666666 # f'{6:6<6}', means start with 6, left align with 6 spaces filled with 6
7777777 # f'{7:7<7}', means start with 7, left align with 7 spaces filled with 7
88888888 # f'{8:8<8}', means start with 8, left align with 8 spaces filled with 8
999999999 # f'{9:9<9}', means start with 9, left align with 9 spaces filled with 9
You already know what the f-string f'{a:{a}<{a}' does - when given an {object} within the string it will replace with said object. In this case, a is range of 1 to 9.
Then all you need to understand is what {9:9<9} does. It's a string formatter as the documentation that the answer has provided:
'<' Forces the field to be left-aligned within the available space (this is the default for most objects).
The x<y portion signifies to left-align the text with a width of y spaces. For any unused space, pad it with character x. So you start with {9} as the first character, and for the rest of the 8 unused spaces, fill it with {9}. This is what {9:9<9} does.
Then you apply the same logic and see how each iteration came about.
More importantly, it should be noted that what feels like "magic" is often just a lack of understanding. Once you take the time to digest and understand the process, it becomes very disillusioning and you become enlightened.