Use hex-encoded or string representation for user-facing identifiers

[G8XSU]

If an identifier needs to be handled, referred to, read, or logged, it should be hex-encoded or in its string representation.

Byte arrays provide a poor user experience in these cases, making tasks like debugging and logging unnecessarily difficult.

While byte arrays are acceptable for machine-handled data (e.g., encrypted bytes, secrets, or protobufs), user-facing data, such as payment-id, channel-id, user-channel-id, and payment-hash, should be more readable.

This applies to both API input and output to reduce friction in interactions with the API.

While someone could easily do this themselves and we can keep on having byte arrays in apis, but it would be nice if clients could log/see responses directly.

[G8XSU]

Consider implementing to_lower_hex() in ldk for ChannelId, PaymentId, OfferId.

[G8XSU]

Consider implementing to_lower_hex() in ldk for ChannelId, PaymentId, OfferId.

Created: lightningdevkit/rust-lightning#3377