Fix non-bijective cipher mapping and add bidirectional encoding

[Copilot]

The English sample cipher had non-bijective mappings causing incorrect decoding (e.g., both 'I' and 'N' mapped to the same ciphertext letter, producing "HY ANY GIVEN STRETHH" instead of "IN ANY GIVEN STRETCH").

Changes

Fixed cipher text mapping

• Replaced sample with proper 1:1 substitution where each plaintext letter maps to exactly one unique ciphertext letter
• Old: YKTJXTFEN QFQSNLOL OL WQLTR GF ZIT YQEZ... (collision: I→O, N→O)
• New: GRTCVTOJD QOQSDYPY PY MQYTZ XO EKT GQJE... (bijective)

Added encode feature

• New textarea accepts plaintext input
• Real-time encoding using inverse of current substitution mapping
• Copy button for encoded output
• Updates automatically when user modifies substitution mappings

// Reverse the substitution mapping for encoding
const reverseSubstitutions = {};
for (const encryptedChar in substitutions) {
    const decodedChar = substitutions[encryptedChar];
    if (decodedChar) {
        reverseSubstitutions[decodedChar] = encryptedChar;
    }
}

Screenshot

Students can now both decode encrypted messages and encode plaintext using their discovered substitutions.

Original prompt

The sample text is not correctly encoded. When I decoded with the frequency analysis tool I get this FREQUENCY ANALYSIS IS BASED ON THE FACT THAT, HY ANY GIVEN STRETHH OF VRHTTEN LANGUAGE, CERTAHN LETTERS AND COMBINATIONS OF LETTERS OCCUR VITH VARYHNG FREQUENCES. MOREOVER, THERE IS A CHARACTERISTHE DISTRIUUTION OF LETTERS THAT IS ROUGHLP THE SAME FOR ALMOST ALL SAMPLSL OF THAT LANGUAGE. I think that more than one letter are assigned to the same. Review the logic and fix it. Then add the hability for the user to select the substitution and show the encoded message.

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.