Reqwest Client (Tokio 1.0 support)

[msrd0]

Fixes #82
Fixes #90

Description

Using reqwest over surf has a few advantages:

• it has already update to tokio 1.0, which has been out there for 6 month now
• it's less of a dependency hell than surf

However, I've kept the existing surf-based feature flags for those that really need async-std compatibily.

Some notes:

• I've bumped the MSRV from 1.45 to 1.46 so that the socket2 dependency compiles
• I've made it a hard error if both surf and reqwest features are enabled. We could prioritize one over the other, however, that would make the #[cfg] statements more complex. Also, there are other error messages than my custom ones, which could be suppressed but would add more complexity.

Checklist

• Formatted code using cargo fmt --all
• Linted code using clippy cargo clippy --all-targets --all-features -- -D warnings - I didn't introduce those warnings
• Updated README.md using cargo readme -r influxdb -t ../README.tpl > README.md
• Reviewed the diff. Did you leave any print statements or unnecessary comments?
• Any unfinished work that warrants a separate issue captured in an issue with a TODO code comment

[msrd0]

I think there's only the <dyn> fixes missing

@Empty2k12 As I mentioned before, these dyn-related warnings have not been introduced by me or this PR. They have been introduced by the Rust compiler and are present in the code on the master branch as well. Since there is no warningless version of the master branch, I had no choice but to use your warning-"infested" code on the master branch as a base for this PR.

I've asked you before if you want me to do a separate PR to fix those warnings, and suggested to just remove the offensive function, as it just calls another public function and nothing else. Your only reply was Gute Frage.

[marcelbuesing]

I think there's only the <dyn> fixes missing

@Empty2k12 As I mentioned before, these dyn-related warnings have not been introduced by me or this PR. They have been introduced by the Rust compiler and are present in the code on the master branch as well. Since there is no warningless version of the master branch, I had no choice but to use your warning-"infested" code on the master branch as a base for this PR.

I've asked you before if you want me to do a separate PR to fix those warnings, and suggested to just remove the offensive function, as it just calls another public function and nothing else. Your only reply was Gute Frage.

If this is the only thing keeping this from being merged I think it might make sense to just add it to the PR. I think tokio 1.x support is essential.

[msrd0]

If this is the only thing keeping this from being merged I think it might make sense to just add it to the PR. I think tokio 1.x support is essential.

If @Empty2k12 could decide if he wants the API to promote <dyn Query>::raw_read_query (which IMHO looks ugly) or replace it by ReadQuery::new (which exists today and does exactly the same) I'm happy to make the required changes.