把图的深度优先遍历、广度优先遍历改为输出深度优先生成树、广度优先生成树代码

深度优先生成树（DFS生成树）的代码：

def dfs_tree(graph, start):
    visited = set()
    tree = {}
    def dfs(node, parent=None):
        visited.add(node)
        if parent:
            if parent not in tree:
                tree[parent] = []
            tree[parent].append(node)
        for neighbor in graph[node]:
            if neighbor not in visited:
                dfs(neighbor, node)
    dfs(start)
    return tree

广度优先生成树（BFS生成树）的代码：

from collections import deque

def bfs_tree(graph, start):
    visited = set()
    tree = {}
    queue = deque([(start, None)])
    while queue:
        node, parent = queue.popleft()
        visited.add(node)
        if parent:
            if parent not in tree:
                tree[parent] = []
            tree[parent].append(node)
        for neighbor in graph[node]:
            if neighbor not in visited:
                queue.append((neighbor, node))
                visited.add(neighbor)
    return tree

其中，`graph` 是图的邻接表表示，`start` 是遍历起点。生成树的结果以字典形式返回，其中键是父节点，值是其直接子节点列表。