Add local fuzz testing setup

[arman-bd]

Add local fuzz testing setup for httpmorph before considering OSS-Fuzz integration

I want to introduce a lightweight fuzzing setup for httpmorph to catch edge-case bugs and improve the library’s robustness before moving toward a full OSS-Fuzz integration.

---

🎯 Goal

Start with a local fuzzing workflow that can run on my machine or CI (e.g., GitHub Actions) without relying on Google’s OSS-Fuzz infrastructure. Once the harnesses are mature and cover key components, I’ll consider submitting httpmorph to OSS-Fuzz for continuous large-scale testing.

---

🧩 Plan

1. Add basic fuzz harnesses using [Atheris](https://github.com/google/atheris) (Python coverage-guided fuzzer) targeting core functions such as:

   • URL parsing and normalization
   • Header parsing and request construction
   • Response parsing (status line, headers, chunked bodies)

2. Run fuzzing locally with:

   python3 -m pip install atheris
   python3 fuzz_parse_request.py

3. Integrate with CI to run short fuzz sessions on pull requests — for example, using GitHub Actions with a time limit (e.g., 60 s per harness).

4. Collect crash inputs and create reproducible test cases to fix discovered issues early.

---

🧠 Benefits

• Detect unexpected exceptions, parsing errors, or logic bugs before they reach users.
• Build a stable foundation for later OSS-Fuzz adoption.
• Keep the process fast and developer-friendly (no complex infrastructure needed yet).

Once I’m confident in the fuzz harness coverage and results, I’ll open a follow-up issue about integrating httpmorph into OSS-Fuzz for continuous large-scale fuzzing.

— Arman ([@arman-bd](https://github.com/arman-bd))