`?` currently influences inference s.t. for `error()?` rustc infers `T = ()`.
However, it is quite confusing -- `?` is not supposed to influence inference,
it's just a conicidence because of the curent implementation. There is an idea
to change this behavior in future Rust versions, such that this code won't
compile without this change.

* .cargo/audit.toml: ignore RUSTSEC-2023-0071 (#1276)

It's not actionable until a new release of the `rsa` crate is available

* der v0.7.9

Changed
- make sure der is comptatible with potential language breaking changed (backport of #1374)
- ignore RUSTSEC-2023-0071 (backport of #1276)

* der-derive: avoid type inference when using default (#1442)

* Revert "x509-cert: specify concrete types to help the compiler (#1441)"

This reverts commit 7a2d38a.

* der-derive: avoid type inference when using default

This is another take on #1441. This is intended to make sure we can
still continue to use `Default::default` and not break future releases
by mistake.

* der-derive: v0.7.3

Changed (2024-07-09)
- avoid type inference when using default (#1443)

* rust 1.78 fixups

This fixes all `unnecessary qualification` warnings

* x509-cert: ignore zlint error, fixed in main

* x509-cert: fuzz: ignore unused_qualification errors

* der: fix append in `Encode::encode_to_vec`

* der: test encode_to_vec with append

Co-authored-by: dishmaker <141624503+dishmaker@users.noreply.github.com>

Fixed:
- append in `Encode::encode_to_vec` (backport [#1760])

[#1760]: #1760