Investigate if RSA issue can happen with ECDSA/ECDH on OpenSSL < 3.0

[krwq]

In #104961 we've changed OpenSSL implementation of ECDsa and ECDH to be similar to RSA. RSA implementation had a workaround for OpenSSL issue which occurs only on some low versions of OpenSSL and it requires us checking if key is a private key explicitly rather than relying on OpenSSL API. See: #53345 (comment) - we've added HasNoPrivateKey check in the Sign/Decrypt operations.

We need to verify if:

• is that code still needed (i.e. has OpenSSL fixed the bug)
• do we need similar check in ECDSA/ECDH? (the most likely answer is "no" but we need to confirm)

As part of this it would be good to add provider test cases as suggested per #104961 (review)

[dotnet-policy-service]

Tagging subscribers to this area: @dotnet/area-system-security, @bartonjs, @vcsjones
See info in area-owners.md if you want to be subscribed.