Bump bytestring to >= 0.10 for correct IsString ByteString instance

[andreasabel]

Bump bytestring to >= 0.10 for correct IsString ByteString instance.
See #13 (comment)

This PR delivers a fix contained (in different form) in PRs:

• bump base16-bytestring bound + add github actions #13
• Allow base16-bytestring-1.0 #10

[emilypi]

LGTM, tho, to follow up, the stats show <2% of users are on GHCs <8.4 as of the last Yearly Haskell Survey in 2021, which is a further diminishing category as time goes on (esp. with respect to the 7.x series). Considering that the lower bound cap for base places you outside of the large number of performance improvements seen in bytestring-0.11, I'm not sure that keeping these old GHC versions is the prudent choice. The rest of the ecosystem, as well as most of the core libraries, now require a GHC of at least 8.x. What do you think about dropping the older GHCs to bring the library in line with what you can reasonably test?

[andreasabel]

I hear your message.
I am happy to drop any GHC 7.x if it brings trouble. E.g., should resolv get new functionality which needs testing or wants to use newer GHC features, there would be no reason to hang on to ancient GHCs. But I'd drop these if there is need; otherwise, while resolv is stable, I see no harm in keeping them around.

[andreasabel]

@phadej: You are hackage maintainer of resolv but had no push-rights on this repo. For me, its the converse: I can admin this repo but no release to hackage.
I rectified this from my side and gave you push rights here.
Basic general question: Do you want to be the maintainer for resolv? Then I leave further processing to your hands.
Otherwise, if you want me to (co-)maintain, please add me to the hackage maintainers.
In this case, there is still the question whether this bound tightening should be a revision or a patch release. I am tending to the latter.