fix some potential audio issues, clean up the code a bit, and suppress some warnings

[xianshijing-lk]

Two changes on the C++ code:

1. one is to use std::vector rather than raw pointer for silence_audio_;
2. when injecting silence to sinks after missing 10 frames, current code will keep injecting and not reset the counter; instead, we should reset the counter once we get real audio, this will prevent breaking the audio.

Suppressing some compilation warnings:
66 | fn utf8_aware_chunks(&self, chunk_size: usize) -> Utf8AwareChunks {
| ^^^^^ --------------- the same lifetime is hidden here
| |
| the lifetime is elided here
|
= help: the same lifetime is referred to in inconsistent ways, making the signature confusing
help: use '_ for type paths
|
66 | fn utf8_aware_chunks(&self, chunk_size: usize) -> Utf8AwareChunks<'_> {
| ++++

warning: hiding a lifetime that's elided elsewhere is confusing

and one webrtc related nullable warning

[CLAassistant]

All committers have signed the CLA.

[ladvoc]

@xianshijing-lk, please merge from main—this will fix CI build.

[cloudwebrtc]

lgtm!

[xianshijing-lk]

@xianshijing-lk, please merge from main—this will fix CI build.

I rebased, but the CI still fails. The logs show that it does not have protoc, any idea what could the problem ?

Protobuf generation failed: Custom { kind: NotFound, error: "Could not find protoc. If protoc is installed, try setting the PROTOC environment variable to the path of the protoc binary. To install it on macOS, run brew install protobuf. It is also available at https://github.com/protocolbuffers/protobuf/releases For more information: https://docs.rs/prost-build/#sourcing-protoc" }