Fix typed data signing for string fields containing hex addresses

[Copilot]

Typed data signing was failing with StarknetTypedDataError.decodingError when string fields contained hex addresses. This issue occurred specifically with typed data like:

{
  "types": {
    "AccountCreation": [
      {"name": "wallet", "type": "string"}
    ]
  },
  "message": {
    "wallet": "0x2a7877881132f27cbfff1021f0ca82ffbcadd14e7599f55e03e891c01edc534"
  }
}

Root Cause:
During JSON decoding, hex strings like "0x2a7877..." were being parsed as Felt elements instead of remaining as String elements due to aggressive number parsing in Element.init(from decoder:). Later, when processing string types in revision v1, the unwrapLongString method expected a .string(String) element but received a .felt(Felt) element, causing the decoding error shown in the stack trace.

Solution:
Modified unwrapLongString(from element:) to handle both .string(String) and .felt(Felt) elements:

• .string(String) elements: processed directly (preserves existing behavior)
• .felt(Felt) elements: converted back to hex string using felt.toHex() method

This approach maintains backward compatibility while fixing the specific case where hex strings were incorrectly parsed during JSON decoding.

Testing:

• Added comprehensive test cases covering the exact issue scenario
• Tested edge cases (short hex, long hex, empty strings, regular text)
• Verified compatibility with both v0 and v1 revisions
• Confirmed existing tests continue to pass (they use regular text strings)

Fixes #251.

---

💬 Share your feedback on Copilot coding agent for the chance to win a $200 gift card! Click here to start the survey.

[Copilot]

@franciszekjob 👋 This repository doesn't have Copilot instructions. With Copilot instructions, I can understand the repository better, work faster and produce higher quality PRs.

I can generate a .github/copilot-instructions.md file for you automatically. Click here to open a pre-filled issue and assign it to me. I'll write the instructions, and then tag you for review.