Minimum Sprints to Complete All Tasks Given Dependencies

Question

## Problem

You have `n` tasks (0 to n-1) and a list of dependencies `[a, b]` meaning task `b` must complete before task `a`. Tasks with no pending dependencies can run in parallel within the same sprint. Find the **minimum number of sprints** to complete all tasks.

## Example
```
n = 6
dependencies = [[2,0],[2,1],[3,1],[4,3],[5,2],[5,3]]
Output: 4
```

## Approach
Topological BFS (Kahn's algorithm). The answer is the length of the longest dependency chain.

## Constraints
- 1 ≤ n ≤ 10^4
- Acyclic graph guaranteed (valid sprint plan exists)

## Atlassian follow-up
What if a task can be split across sprints? What if certain tasks have a fixed sprint they must start in?