Added log for instruction counts in stwo-adapted-prover

[yuvalsw]

This change is 

[yuvalsw]

change target branch

[Alon-Ti]

Reviewed 3 of 4 files at r1, 1 of 1 files at r2, all commit messages.
Reviewable status: all files reviewed, 6 unresolved discussions (waiting on @yuvalsw)

---

stwo_cairo_prover/crates/adapted_prover/src/main.rs line 67 at r2 (raw file):

    let instruction_counts = vm_output.instructions.counts();
    log::info!("Instruction counts: {instruction_counts:?}");

Is it better to impl Display for this so you don't have to use :??

---

stwo_cairo_prover/crates/prover/src/input/instructions.rs line 27 at r2 (raw file):

///
/// For each opcode with flags, the array describes the different flag combinations. The index
/// refers to the flag combination in little endian. For example, jnz_imm at index 1 (100 in little

bit-reversed

Code quote:

little endian

---

stwo_cairo_prover/crates/prover/src/input/instructions.rs line 27 at r2 (raw file):

///
/// For each opcode with flags, the array describes the different flag combinations. The index
/// refers to the flag combination in little endian. For example, jnz_imm at index 1 (100 in little

Maybe add a TODO to make this indexing mechanism more explicit in code?

---

stwo_cairo_prover/crates/prover/src/input/instructions.rs line 29 at r2 (raw file):

/// refers to the flag combination in little endian. For example, jnz_imm at index 1 (100 in little
/// endian) is for: fp (1=true), not taken (0=false), no ap++ (0=false).
/// Note: for the flag "fp/ap", true means fp-based and false means ap-based.

If this is the case you should keep the ap/fp order in the comments.

---

stwo_cairo_prover/crates/prover/src/input/instructions.rs line 352 at r2 (raw file):

/// The counts of the instructions usage in the input, split to Stwo opcodes.
///
/// For each opcode with flags, the array describes the different flag combinations. The index

Replace repeated comment with "as above" (or, ideally, share the code with the above struct with generics so that neither the comments nor the vector sizes have to be repeated).