import re

text = ""
with open("read.html", "r") as rf:
    text = rf.read()
    
pattern = r"<link[\s\S]*?<link"
results = re.findall(pattern, text)
if results:
    r = results[0]
    with open("write.html", "w") as wf:
        wf.write(r)
            
================================================

with open("read.html", "r") as rf:
    with open("write.html", "w") as wf:
        num = 0
        for line in rf.readlines():
            if line.startswith("<link"):
                num += 1
                continue
            if num == 2:
                break
            wf.writelines(line)