scrapy gunzip inclusion?

[nyov]

Would scrapy's gunzip and is_gzipped functions (scrapy/utils/gz.py) fit in w3lib?
If is_gzipped would take headers instead of response as argument, there is no other dependency.

[Gallaecio]

I’ve just proposed to deprecate is_gzipped in Scrapy: scrapy/scrapy#3718

Maybe we could implement Scrapy’s gzip_magic_number as is_gzipped here?

As for gunzip, it may make sense to move it to w3lib. I think it would become the easiest answer for https://stackoverflow.com/q/1732709/939364

@redapple, @dangra, @kmike Thoughts?