fix: use usize instead of u64 for batch-size arg

[KevFan]

Limitador Operator using the latest limitador image with redis has a failing test when starting up the limitador pod with the following error since #318 was introduced:

➜  ~ kubectl logs limitador-limitador-lcrp7-dffc56ff-qzccn -n test-namespace-8sgz7
thread 'main' panicked at /usr/local/cargo/registry/src/index.crates.io-6f17d22bba15001f/clap_builder-4.5.2/src/parser/error.rs:32:9:
Mismatch between definition and access of `batch`. Could not downcast to TypeId { t: (8519994227001858441, 10522819541147869382) }, need to downcast to TypeId { t: (11446210613632762899, 3222440509213045925) }

note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace

Seems to be related to mismatch of types defined at:

limitador/limitador-server/src/config.rs

Line 167 in 25346a6

pub batch_size: usize,

vs

limitador/limitador-server/src/main.rs

Line 596 in 25346a6

.value_parser(clap::value_parser!(u64))

Caveat: Not sure is this the correct change since I don't know Rust but changing the type and using a custom image with the change passed the failing test in limitador operator at least 🤷

[alexsnaps]

Thanks for catching this @KevFan
This whole i vs. u and 64 vs size are something I'd really like us to clean up entirely
i64 is abused afaict because that's the type you get from redis...

[didierofrivia]

🔎