chore: benchmark sha256 number of instructions executed in AVM

[dbanks12]

Also move hash stuff into another AVM test contract.

This lets us get stats like below, although note that this isn't directly an output from the yarn tests:

For a sha256 of N bytes, how many AVM/Brillig instructions are executed and how much gas does it consume?

N=  20 bytes:   4513 instructions,   77778 L2 Gas
N=  30 bytes:   5328 instructions,   91326 L2 Gas
N=  40 bytes:   5843 instructions,  100098 L2 Gas
N=  50 bytes:   6672 instructions,  113877 L2 Gas
N=  60 bytes:   7490 instructions,  127626 L2 Gas
N=  70 bytes:   8662 instructions,  147936 L2 Gas
N=  80 bytes:   9207 instructions,  157368 L2 Gas
N=  90 bytes:  10052 instructions,  171576 L2 Gas
N= 100 bytes:  10597 instructions,  181008 L2 Gas
N= 255 bytes:  23046 instructions,  392055 L2 Gas
N= 256 bytes:  23022 instructions,  392121 L2 Gas
N= 511 bytes:  43107 instructions,  732336 L2 Gas
N= 512 bytes:  42978 instructions,  731004 L2 Gas
N=2048 bytes: 162801 instructions, 2765820 L2 Gas```

[github-actions]

Changes to public function bytecode sizes

Generated at commit: ff8871acf94d03e1927675f41661fa062b0009c0, compared to commit: 2ba1e7112c4d8052967603ab78c1213cc70b8038

🧾 Summary (100% most significant diffs)

Program	Bytecode size in bytes (+/-)	%
AvmTest::public_dispatch	+6,647 ❌	+10.55%

---

Full diff report 👇

Program	Bytecode size in bytes (+/-)	%
AvmTest::public_dispatch	69,677 (+6,647)	+10.55%

[jeanmon]

@dbanks12 I would directly add this code in a bash script and commit it to the repo.