fix(python): add model aliases to API + graceful port conflict handling

1. Model aliases in Python API (#74):
   - `Model.from_pretrained("llama3.2:1b")` now works (was: only full names)
   - Added `_MODEL_ALIASES` dict and `_resolve_model_name()` in __init__.py
   - Same aliases as CLI: smollm2, llama3.2:1b, phi3.5:mini, etc.

2. Port conflict handling (#75):
   - `quantcpp serve` now checks if port is in use before launching
   - Shows: "error: port 8080 is already in use. Try --port 8081"
   - Previously: SIGABRT crash with no error message

Fixes #74, #75

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: unamedkr

bindings/python/quantcpp/__init__.py

@@ -155,14 +155,36 @@ def _download_with_progress(url: str, dest: Path, desc: str) -> None:
     tmp.rename(dest)
 
 
+_MODEL_ALIASES = {
+    "smollm2":         "SmolLM2-1.7B",
+    "smollm2:1.7b":    "SmolLM2-1.7B",
+    "smollm2:135m":    "SmolLM2-135M",
+    "qwen3.5":         "Qwen3.5-0.8B",
+    "qwen3.5:0.8b":    "Qwen3.5-0.8B",
+    "llama3.2":        "Llama-3.2-1B",
+    "llama3.2:1b":     "Llama-3.2-1B",
+    "phi3.5":          "Phi-3.5-mini",
+    "phi3.5:mini":     "Phi-3.5-mini",
+    "phi-3.5":         "Phi-3.5-mini",
+    "phi-3.5-mini":    "Phi-3.5-mini",
+}
+
+
+def _resolve_model_name(name: str) -> str:
+    """Resolve alias or case-insensitive name to canonical registry key."""
+    if name in _MODEL_REGISTRY:
+        return name
+    return _MODEL_ALIASES.get(name.lower(), name)
+
+
 def download(name: str) -> str:
     """Download a model from HuggingFace Hub and return its local path.
 
     Parameters
     ----------
     name : str
-        Model name from the registry. Currently available:
-        ``"SmolLM2-135M"`` (~135 MB, good for testing).
+        Model name or alias. Examples: ``"Phi-3.5-mini"``, ``"phi3.5:mini"``,
+        ``"smollm2"``, ``"llama3.2:1b"``.
 
     Returns
     -------
@@ -171,9 +193,10 @@ def download(name: str) -> str:
 
     Examples
     --------
-    >>> path = quantcpp.download("SmolLM2-135M")
+    >>> path = quantcpp.download("phi3.5:mini")
     >>> m = quantcpp.Model(path)
     """
+    name = _resolve_model_name(name)
     if name not in _MODEL_REGISTRY:
         avail = ", ".join(sorted(_MODEL_REGISTRY))
         raise ValueError(

bindings/python/quantcpp/cli.py

@@ -250,6 +250,15 @@ def cmd_serve(args):
         print("  Or install via your package manager.", file=sys.stderr)
         return 2
 
+    # Check if port is available before launching server
+    import socket
+    with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as sock:
+        if sock.connect_ex(("127.0.0.1", args.port)) == 0:
+            print(f"error: port {args.port} is already in use.", file=sys.stderr)
+            print(f"  Try a different port: quantcpp serve {args.model} --port {args.port + 1}",
+                  file=sys.stderr)
+            return 1
+
     cmd = [binary, model_path, "-p", str(args.port), "-j", str(args.threads)]
     print(f"quantcpp serve {os.path.basename(model_path)} on :{args.port}", file=sys.stderr)
     print("", file=sys.stderr)