Rename module to sneak.berlin/go/x and rework bip39wordlist

Move to Go 1.26, embed the wordlist from english.txt, add
ValidateWordlist, NormalizeWordlist, and GenerateMnemonic with tests,
and remove the old root-level CLI that used the previous module path.

pkg/bip39wordlist/wordlist.go

@@ -0,0 +1,42 @@
+// Package bip39wordlist provides the BIP39 English wordlist and helpers for
+// validating, normalizing, and generating BIP39 mnemonic phrases.
+//
+// The wordlist is embedded in the binary from english.txt and parsed once
+// at package initialization. The mnemonic helpers accept phrases in any
+// letter case and with any whitespace between words; NormalizeWordlist
+// returns the canonical lowercase, single-space-separated form that BIP39
+// seed derivation requires, and refuses anything that is not a valid
+// mnemonic.
+package bip39wordlist
+
+import (
+	_ "embed"
+	"slices"
+	"strings"
+)
+
+// WordlistSize is the number of words in a BIP39 wordlist.
+const WordlistSize = 2048
+
+//go:embed english.txt
+var englishWordlistData string
+
+// WordlistEnglish is the BIP39 English wordlist, one entry per word, in
+// the canonical (sorted) BIP39 order. Callers must not modify it.
+//
+//nolint:gochecknoglobals // static lookup table parsed once from embedded data
+var WordlistEnglish = strings.Fields(englishWordlistData)
+
+// Wordlist is the default wordlist, which is WordlistEnglish. Callers must
+// not modify it.
+//
+//nolint:gochecknoglobals // static lookup table, alias of WordlistEnglish
+var Wordlist = WordlistEnglish
+
+// Contains reports whether word is in the English wordlist. The comparison
+// is exact, so word must already be lowercase.
+func Contains(word string) bool {
+	_, found := slices.BinarySearch(WordlistEnglish, word)
+
+	return found
+}