https://quicwg.org/base-drafts/draft-ietf-quic-transport.html#section-14 only says that padding is done by PADDING frame. https://quicwg.org/base-drafts/draft-ietf-quic-transport.html#section-14

But it looks like it is a common practice to add 0 or garbage outside QUIC packet to fill remaining UDP packet payload to expand packet size.
If it is a good practice (and/or recommended), I think draft should say something about it.
It might slightly simplify the implementation.
Padding in this way has different properties than PADDING frame; for example, it does not contribute to in-flight bytes.