fix(instrumentation-http): add http.host attribute before sending the request

[cuichenli]

Which problem is this PR solving?

Instead of add the http.host attribute after received the response, this PR changed it to add the attribute before the request was sent.

Fixes #3031

Short description of the changes

Infer the hostname and port from the host field

Type of change

Please delete options that are not relevant.

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

How Has This Been Tested?

Unit test

[codecov]

Codecov Report

Merging #3054 (a79e9f4) into main (e76cb8b) will increase coverage by 0.01%.
The diff coverage is 100.00%.

❗ Current head a79e9f4 differs from pull request most recent head 5204342. Consider uploading reports for the commit 5204342 to get more accurate results

@@            Coverage Diff             @@
##             main    #3054      +/-   ##
==========================================
+ Coverage   92.69%   92.70%   +0.01%     
==========================================
  Files         181      181              
  Lines        5655     5664       +9     
  Branches     1201     1202       +1     
==========================================
+ Hits         5242     5251       +9     
  Misses        413      413              

Impacted Files	Coverage Δ
...ges/opentelemetry-instrumentation-http/src/http.ts	94.42% <100.00%> (-0.04%)	⬇️
...es/opentelemetry-instrumentation-http/src/utils.ts	99.06% <100.00%> (+0.05%)	⬆️

[vmarchaud]

could you rebase your PR ? otherwise the small nit, lgtm

[legendecas]

The attribute changes LGTM. However, I find that the new changes on the timeout event are not convincing. Would you mind splitting that into another PR?

[legendecas]

Thank you for your contribution!