Make unicode normalization optional

[elsirion]

Potential solution for #28 (comment)

With default features this is the dependency tree now:

bip39 v2.1.0 (/home/user/projects/elsirion/rust-bip39)
├── bitcoin_hashes v0.13.0
│   ├── bitcoin-internals v0.2.0
│   └── hex-conservative v0.1.2
├── rand v0.8.5
│   ├── libc v0.2.162
│   ├── rand_chacha v0.3.1
│   │   ├── ppv-lite86 v0.2.20
│   │   │   └── zerocopy v0.7.35
│   │   │       ├── byteorder v1.5.0
│   │   │       └── zerocopy-derive v0.7.35 (proc-macro)
│   │   │           ├── proc-macro2 v1.0.89
│   │   │           │   └── unicode-ident v1.0.13
│   │   │           ├── quote v1.0.37
│   │   │           │   └── proc-macro2 v1.0.89 (*)
│   │   │           └── syn v2.0.87
│   │   │               ├── proc-macro2 v1.0.89 (*)
│   │   │               ├── quote v1.0.37 (*)
│   │   │               └── unicode-ident v1.0.13
│   │   └── rand_core v0.6.4
│   │       └── getrandom v0.2.15
│   │           ├── cfg-if v1.0.0
│   │           └── libc v0.2.162
│   └── rand_core v0.6.4 (*)
├── rand_core v0.6.4 (*)
└── serde v1.0.214

[elsirion]

Idk if that's the right way to go about it, but it seems reasonable to want UN if you are dealing with these languages. I can also make it an entirely optional feature, which is cleaner but imo risks misuse.

[maan2003]

doesn't work in old rust

[dpc]

@stevenroose What was the ancient Rust version we need to get it to work on? :D

[dpc]

@elsirion I guess we don't really need to use dep: which is somehow recent addition.

And we can have unicode-normalization-std to avoid using ?.