Use our own custom badssl.julialang.org implementation

[DilumAluthge]

Follow-up to #176 and #177.

[DilumAluthge]

@staticfloat @StefanKarpinski We now have some evidence suggesting that the public badssl.com instance was not the cause of the Base Julia CI failures. More details in JuliaLang/julia#44123.

So, while we still want to ultimately move to using our own instance, there is now no longer any rush on that. So I would say that @staticfloat can take his time working on a custom implementation for these two endpoints (untrusted-root and wrong.host), and then we can wait until that work is done before we merge this PR.

[codecov]

Codecov Report

Merging #178 (53bb556) into master (a41840c) will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master     #178   +/-   ##
=======================================
  Coverage   92.47%   92.47%           
=======================================
  Files           5        5           
  Lines         518      518           
=======================================
  Hits          479      479           
  Misses         39       39           

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update a41840c...53bb556. Read the comment docs.

[DilumAluthge]

We now have our own custom instance of badssl:

1. https://github.com/staticfloat/badssl.julialang.org
2. badssl.julialang.org
   • https://badssl.julialang.org/
   • https://right.host.badssl.julialang.org/
   • https://wrong.host.badssl.julialang.org/
   • https://untrusted-root.badssl.julialang.org/

Therefore, this PR is now ready to merge (assuming that all CI jobs pass).

[DilumAluthge]

• 🤖 Bump the Downloads stdlib from 43899e7 to 2a21b15 julia#44148