Add "algorithm mismatch" error to improve jws

[ghost]

Upstream libraries that depend on jws.verify() break when the upstream keys contain a mixed set of algorithms. This is a nominal occurance for OIDC servers and should be properly handled.

[codecov]

Codecov Report

Merging #304 (546e96c) into master (96474ec) will decrease coverage by 0.11%.
The diff coverage is 84.00%.

❗ Current head 546e96c differs from pull request most recent head 1ce256e. Consider uploading reports for the commit 1ce256e to get more accurate results

@@            Coverage Diff             @@
##           master     #304      +/-   ##
==========================================
- Coverage   92.94%   92.83%   -0.12%     
==========================================
  Files          15       15              
  Lines        1418     1423       +5     
==========================================
+ Hits         1318     1321       +3     
- Misses        100      102       +2     

Impacted Files	Coverage Δ
jose/jws.py	93.54% <60.00%> (-1.50%)	⬇️
jose/backends/cryptography_backend.py	93.18% <77.77%> (ø)
jose/backends/ecdsa_backend.py	97.50% <100.00%> (ø)
jose/backends/native.py	97.56% <100.00%> (ø)
jose/backends/rsa_backend.py	95.62% <100.00%> (ø)
jose/exceptions.py	100.00% <100.00%> (ø)

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[ghost]

As mentioned in the issue, this implements step 2 of Appendix D of the JWS spec

2. Filter the set of collected keys. For instance, some
   applications will use only keys referenced by "kid" (key ID) or
   "x5t" (X.509 certificate SHA-1 thumbprint) parameters. If the
   application uses the JWK "alg" (algorithm), "use" (public key
   use), or "key_ops" (key operations) parameters, keys with
   inappropriate values of those parameters would be excluded.
   Additionally, keys might be filtered to include or exclude keys
   with certain other member values in an application-specific
   manner. For some applications, no filtering will be applied.

[dimaqq]

@mpdavis given that the original PR author account is deleted, maybe it's time to make a call: either take this PR over, maybe add more tests and merge it, or close it if it's incomplete?

My 2c: this PR is a good start.