Add seeded task workspace creation

Current persistent tasks started with an empty workspace, which blocked the first useful host-to-task workflow in the task roadmap. This change lets task creation start from a host directory or tar archive without changing the one-shot VM surfaces.

Expose source_path on task create across the CLI, SDK, and MCP, add safe archive upload and extraction support for guest and host-compat backends, persist workspace_seed metadata, and patch the per-task rootfs with the bundled guest agent before boot so seeded guest tasks work without republishing environments. Also switch post--- command reconstruction to shlex.join() so documented sh -lc task examples preserve argument boundaries.

Validation:
- uv lock
- UV_CACHE_DIR=.uv-cache uv run pytest --no-cov tests/test_vm_guest.py tests/test_vm_manager.py tests/test_cli.py tests/test_api.py tests/test_server.py tests/test_public_contract.py
- UV_CACHE_DIR=.uv-cache make check
- UV_CACHE_DIR=.uv-cache make dist-check
- real guest-backed smoke: task create --source-path, task exec -- cat note.txt, task delete

src/pyro_mcp/cli.py

@@ -4,6 +4,7 @@ from __future__ import annotations
 
 import argparse
 import json
+import shlex
 import sys
 from textwrap import dedent
 from typing import Any
@@ -155,6 +156,14 @@ def _print_task_summary_human(payload: dict[str, Any], *, action: str) -> None:
     print(f"Environment: {str(payload.get('environment', 'unknown'))}")
     print(f"State: {str(payload.get('state', 'unknown'))}")
     print(f"Workspace: {str(payload.get('workspace_path', '/workspace'))}")
+    workspace_seed = payload.get("workspace_seed")
+    if isinstance(workspace_seed, dict):
+        mode = str(workspace_seed.get("mode", "empty"))
+        source_path = workspace_seed.get("source_path")
+        if isinstance(source_path, str) and source_path != "":
+            print(f"Workspace seed: {mode} from {source_path}")
+        else:
+            print(f"Workspace seed: {mode}")
     print(f"Execution mode: {str(payload.get('execution_mode', 'pending'))}")
     print(
         f"Resources: {int(payload.get('vcpu_count', 0))} vCPU / "
@@ -446,7 +455,7 @@ def _build_parser() -> argparse.ArgumentParser:
         epilog=dedent(
             """
             Examples:
-              pyro task create debian:12
+              pyro task create debian:12 --source-path ./repo
               pyro task exec TASK_ID -- sh -lc 'printf "hello\\n" > note.txt'
               pyro task logs TASK_ID
             """
@@ -458,7 +467,13 @@ def _build_parser() -> argparse.ArgumentParser:
         "create",
         help="Create and start a persistent task workspace.",
         description="Create a task workspace that stays alive across repeated exec calls.",
-        epilog="Example:\n  pyro task create debian:12",
+        epilog=dedent(
+            """
+            Examples:
+              pyro task create debian:12
+              pyro task create debian:12 --source-path ./repo
+            """
+        ),
         formatter_class=_HelpFormatter,
     )
     task_create_parser.add_argument(
@@ -497,6 +512,13 @@ def _build_parser() -> argparse.ArgumentParser:
             "is unavailable."
         ),
     )
+    task_create_parser.add_argument(
+        "--source-path",
+        help=(
+            "Optional host directory or .tar/.tar.gz/.tgz archive to seed into `/workspace` "
+            "before the task is returned."
+        ),
+    )
     task_create_parser.add_argument(
         "--json",
         action="store_true",
@@ -663,7 +685,7 @@ def _require_command(command_args: list[str]) -> str:
         command_args = command_args[1:]
     if not command_args:
         raise ValueError("command is required after `--`")
-    return " ".join(command_args)
+    return shlex.join(command_args)
 
 
 def main() -> None:
@@ -764,6 +786,7 @@ def main() -> None:
                 ttl_seconds=args.ttl_seconds,
                 network=args.network,
                 allow_host_compat=args.allow_host_compat,
+                source_path=args.source_path,
             )
             if bool(args.json):
                 _print_json(payload)