unicode/fsm: delete all of the old DFAs

These DFAs were generated for regex-automata 0.1 by ucd-generate. We're
going to start completely fresh with regex-automata 0.3 using regex-cli.

This also lets us drop the once_cell dependency, since we'll plan to use
regex-automata's `Lazy` type for loading DFAs.

Cargo.toml

@@ -25,12 +25,11 @@ bench = false
 default = ["std", "unicode"]
 std = ["alloc", "memchr/std", "serde?/std"]
 alloc = ["serde?/alloc"]
-unicode = ["dep:once_cell", "dep:regex-automata"]
+unicode = ["dep:regex-automata"]
 serde = ["dep:serde"]
 
 [dependencies]
 memchr = { version = "2.4.0", default-features = false }
-once_cell = { version = "1.14.0", optional = true }
 regex-automata = { version = "0.1.5", default-features = false, optional = true }
 serde = { version = "1.0.85", default-features = false, optional = true }
 

src/lib.rs

@@ -394,20 +394,6 @@ and Unicode support.
 #![cfg_attr(not(any(feature = "std", test)), no_std)]
 #![cfg_attr(docsrs, feature(doc_auto_cfg))]
 
-// Why do we do this? Well, in order for us to use once_cell's 'Lazy' type to
-// load DFAs, it requires enabling its 'std' feature. Yet, there is really
-// nothing about our 'unicode' feature that requires 'std'. We could declare
-// that 'unicode = [std, ...]', which would be fine, but once regex-automata
-// 0.3 is a thing, I believe we can drop once_cell altogether and thus drop
-// the need for 'std' to be enabled when 'unicode' is enabled. But if we make
-// 'unicode' also enable 'std', then it would be a breaking change to remove
-// 'std' from that list.
-//
-// So, for right now, we force folks to explicitly say they want 'std' if they
-// want 'unicode'. In the future, we should be able to relax this.
-#[cfg(all(feature = "unicode", not(feature = "std")))]
-compile_error!("enabling 'unicode' requires enabling 'std'");
-
 #[cfg(feature = "alloc")]
 extern crate alloc;