v0.4.5

This patch release includes some fixes to edge conditions in voice receive, and prevents a panic when creating HLS streams in non-tokio contexts. In addition, users of voice receive can now configure the specification of decoded audio (sample rate, channel count) across Opus's supported values.

Thanks to the following for their contributions:

• @FelixMcFelix
• @Erk-

Added

• Receive: Config of decode sample rate/channels (@FelixMcFelix) c:91bf153

Fixed

• Receive: Fix handling for large timestamp/sequence jumps (@FelixMcFelix) c:af95296
• Fix(source): HLS would not work if used directly. (@Erk-) c:312799d