Don't hardcode OPENSSL_ROOT_DIR to /usr on Linux

[SWilson4]

Recent versions of CMake search only OPENSSL_ROOT_DIR if defined, so if OpenSSL is installed in a location other than /usr (e.g., /usr/lib/x86_64-linux-gnu), the build will fail.

Fixes #1748.

• Does this PR change the input/output behaviour of a cryptographic algorithm (i.e., does it change known answer test values)? (If so, a version bump will be required from x.y.z to x.(y+1).0.)
• Does this PR change the list of algorithms available -- either adding, removing, or renaming? Does this PR otherwise change an API? (If so, PRs in fully supported downstream projects dependent on these, i.e., oqs-provider will also need to be ready for review and merge by the time this is merged.)

[baentsch]

Thanks for this improvement @SWilson4 . Conceptually LGTM. Don't understand why we did this hardcoding. As this was 4 years ago, most likely no negative impact to downstreams, but a "trigger-downstream" tag would be good, no?

[SWilson4]

Thanks for this improvement @SWilson4 . Conceptually LGTM. Don't understand why we did this hardcoding. As this was 4 years ago, most likely no negative impact to downstreams, but a "trigger-downstream" tag would be good, no?

It never hurts :)

[SWilson4]

Pinging @open-quantum-safe/liboqs-committers for a second review here.

[SWilson4]

Given that the Travis tests passed on this branch previously, and the only commit since then is the empty downstream trigger, I think it's safe to say this change will not break Travis whenever that CI system is working again (created #1888 to track). Hence I will go ahead and merge.