Add support for secp256k1

[sosthene-nitrokey]

Deppends on:

• Add secp256k1 to mechanism enum trussed-dev/trussed#176
• Add support for secp256k1 trussed-se050-backend#41
• Add support for secp256k1 opcard-rs#213

Closes #531

[nitrokey-ci]

No significant changes.

Insignifcant changes

metric	value		change
binary-size-nk3am	1,371,692	🔴	+4,876 (+0.36%)
binary-size-nk3am-test	1,878,443	🔴	+1,035 (+0.06%)
binary-size-nk3xn	493,496	🔴	+1,632 (+0.33%)
binary-size-nk3xn-test	540,524	🔴	+1,396 (+0.26%)
binary-size-nkpk	710,095	🔴	+445 (+0.06%)

[robin-nitrokey]

What’s going on with the binary size? NKPK is especially strange as it should not be affected by an opcard/SE050 change, but even for the others I would have expected less.

[sosthene-nitrokey]

I think it's because of the bump of toolchain version.

[sosthene-nitrokey]

Oh wait no this is wrong indeed. The metrics are the wrong way, and with the force push they are up-to-date with regards to main.

[sosthene-nitrokey]

It's because of the removal of default-mechanisms which is not merged in trussed upstream but is in our branch.

[robin-nitrokey]

At least a part of the problem is that the trussed commit is based on upstream and not on the fork, so it misses some size optimizations. Not sure if this explains everything.

[sosthene-nitrokey]

Local testing suggest it is the main reason.

[sosthene-nitrokey]

Binary size changes are now insignificant.