chore: improve note_getter performance on >1 notes

[nventuro]

Explanation in the source code.

I've not run our own benchmarks on this yet, but a benchmark from externals reported a decrease from 140k gates to 90k when running a simple contract that read 8 notes, and 260k to 160k with 16 notes. We should see similar results in our own setup.

edit: alternative explanation from Slack, in case anyone finds it useful:

[AztecBot]

Flakey Tests

🤖 says: This CI run detected 2 tests that failed, but were tolerated due to a .test_patterns.yml entry.

\033FLAKED\033 (8;;http://ci.aztec-labs.com/824d08d1272a0921�824d08d1272a09218;;�):  yarn-project/end-to-end/scripts/run_test.sh simple src/e2e_epochs/epochs_high_tps_block_building.test.ts (222s) (code: 1) group:e2e-p2p-epoch-flakes
\033FLAKED\033 (8;;http://ci.aztec-labs.com/091d6450618f551a�091d6450618f551a8;;�): yarn-project/scripts/run_test.sh p2p/src/client/test/p2p_client.integration_message_propagation.test.ts (58s) (code: 1) group:e2e-p2p-epoch-flakes