Add encode/decode functionality to hints.

[yair-starkware]

Issues I found:

1. BigIntAsHex doesn’t implement Encode/Decode and is defined in another crate, so I couldn't implement it myself. Because of this, structs that contain BigIntAsHex also can’t derive Encode/Decode, so I had to implement it manually.
2. Scale encoding ignores field names. For example, in the following code A & B will have the same encoding:

struct A(u8);
struct B{x: 8};

It is dangerous since we won’t be able to detect things like changing fields order, removing + adding fields, etc.
3. Adding option fields breaks the decoding, so we can't rely on it for adding data without migration:

struct A(u8);
struct B(u8, Option<u8>);
let a = A(1).encode();
B::decode(&mut a.as_slice()).unwrap(); // Panics

---

This change is 

[spapinistarkware]

Reviewed 2 of 6 files at r1.
Reviewable status: 2 of 6 files reviewed, 3 unresolved discussions (waiting on @dan-starkware and @yair-starkware)

---

Cargo.toml line 71 at r1 (raw file):

num-traits = "0.2"
once_cell = "1.17.1"
parity-scale-codec = "3.5.0"

Depending on scale is one thing, but are we sure we want to depend on partiy code? Can you explain what exactly this crate has that we need? Regular SCALE doesn't have derive?

Code quote:

parity-scale-codec

---

crates/cairo-lang-casm/src/operand.rs line 63 at r1 (raw file):

    }
}
// Can't derive the trait because BigIntAsHex doesn't implement Encode.

Why not implement it for BigIntAsHex insteaD?

---

crates/cairo-lang-casm/src/hints/mod.rs line 14 at r1 (raw file):

// Represents a cairo hint.
// Note: some projects store the encoding of hints. Be mindful about making changes as they are hard
// to detect and may require storage migration.

.

Suggestion:

// Note: Hint encoding should be backwards-compatible. This is an API guarantee.
// For example, new variants should haev new `index`.

[yair-starkware]

Reviewable status: 2 of 9 files reviewed, 3 unresolved discussions (waiting on @dan-starkware and @spapinistarkware)

---

Cargo.toml line 71 at r1 (raw file):

Previously, spapinistarkware (Shahar Papini) wrote…

Depending on scale is one thing, but are we sure we want to depend on partiy code? Can you explain what exactly this crate has that we need? Regular SCALE doesn't have derive?

WDYM by regular scale?
There are multiple scale related crates, all depending on parity-scale-codec in their basis, if I understand correctly.

---

crates/cairo-lang-casm/src/operand.rs line 63 at r1 (raw file):

Previously, spapinistarkware (Shahar Papini) wrote…

Why not implement it for BigIntAsHex insteaD?

Done, thanks.

---

crates/cairo-lang-casm/src/hints/mod.rs line 14 at r1 (raw file):

Previously, spapinistarkware (Shahar Papini) wrote…

.

Done.

[spapinistarkware]

Reviewed 6 of 7 files at r2, 1 of 1 files at r3, all commit messages.
Reviewable status: complete! all files reviewed, all discussions resolved (waiting on @dan-starkware)

[spapinistarkware]

Reviewed 3 of 3 files at r4, all commit messages.
Reviewable status: complete! all files reviewed, all discussions resolved (waiting on @dan-starkware and @orizi)