feat(cloud): add Vast.ai and Modal GPU providers

Vast.ai (~$0.17/hr A10): SSH+rsync marketplace model with full CLI
(list, launch, terminate, train) matching lambda_labs.py pattern.
Includes GPU search, --gpu-wait retry, auto-convert --demo-dir flow.

Modal ($30/mo free, $1.10/hr A10G): Python-native cloud with zero-ops
training via decorated functions and Modal Volumes for data transfer.
CLI: train, status, download, list-volumes.

Both support the same --demo-dir end-to-end pipeline as Lambda Labs.

53 new tests (34 Vast.ai + 19 Modal), all passing.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Author: abrichr

.env.example

@@ -14,6 +14,37 @@ OPENAI_API_KEY=your-openai-api-key-here
 #
 LAMBDA_API_KEY=your-lambda-api-key-here
 
+# =============================================================================
+# Vast.ai (Cloud GPU Marketplace — cheapest GPUs)
+# =============================================================================
+# Get your API key at: https://cloud.vast.ai/api/
+#
+# Usage:
+#   python -m openadapt_ml.cloud.vast_ai list       # See available GPUs
+#   python -m openadapt_ml.cloud.vast_ai launch      # Launch cheapest A10
+#   python -m openadapt_ml.cloud.vast_ai terminate <id>  # Stop billing!
+#
+VAST_API_KEY=your-vast-api-key-here
+
+# =============================================================================
+# Modal (Serverless Cloud GPU for training)
+# =============================================================================
+# Modal is a Python-native serverless platform with per-second billing.
+# $30/month free credits. No SSH or instances to manage.
+#
+# Setup:
+#   pip install modal
+#   modal token set       # Authenticate (opens browser)
+#
+# Usage:
+#   python -m openadapt_ml.cloud.modal_cloud train --bundle /path/to/bundle
+#   python -m openadapt_ml.cloud.modal_cloud status
+#   python -m openadapt_ml.cloud.modal_cloud download
+#
+# Auth is usually via `modal token set` CLI. Alternatively, set these env vars:
+# MODAL_TOKEN_ID=your-modal-token-id
+# MODAL_TOKEN_SECRET=your-modal-token-secret
+
 # Google Gemini API (for GeminiGrounder element detection)
 # Get your key via one of these methods:
 #

openadapt_ml/cloud/__init__.py

@@ -1,5 +1,6 @@
 """Cloud GPU providers for training."""
 
 from openadapt_ml.cloud.lambda_labs import LambdaLabsClient
+from openadapt_ml.cloud.vast_ai import VastAIClient
 
-__all__ = ["LambdaLabsClient"]
+__all__ = ["LambdaLabsClient", "VastAIClient"]