⚙️ FEATURE-#173: Add dotflow cloud CLI for cross-cloud infrastructure generation

[FernandoCelmer]

Description

Cloud CLI

• dotflow/cli/commands/cloud.py — CloudGenerateCommand and CloudListCommand using local template clone
• dotflow/cli/commands/deploy.py — DeployCommand routing to platform-specific deployers
• dotflow/cli/setup.py — Register cloud and deploy subcommands

Cloud Deploy Module (dotflow/cloud/)

• core.py — ABCs: Registry, RoleManager, LogManager, ObjectStorage, Queue, EventScheduler, Deployer
• aws/base_lambda_deployer.py — Shared logic for all Lambda variants (init, setup, create/update function)
• aws/lambda_deployer.py — Lambda + EventBridge schedule trigger
• aws/lambda_api_deployer.py — Lambda + API Gateway (HTTP API)
• aws/lambda_s3_deployer.py — Lambda + S3 notification trigger
• aws/lambda_sqs_deployer.py — Lambda + SQS event source mapping
• aws/ecs_deployer.py — ECS Fargate task definition + cluster
• aws/constants.py — AWS error messages
• aws/services/ecr.py — ECR registry (Registry ABC)
• aws/services/iam.py — IAM roles (RoleManager ABC)
• aws/services/cloudwatch.py — CloudWatch log groups (LogManager ABC)
• aws/services/s3.py — S3 object storage (ObjectStorage ABC)
• gcp/cloudrun_deployer.py — Cloud Run deployer (Cloud Build + Run SDK)
• gcp/services/gcs.py — GCS object storage (ObjectStorage ABC)
• gcp/services/artifact_registry.py — Artifact Registry (Registry ABC)
• gcp/services/apis.py — GCP API enablement via Service Usage SDK
• gcp/constants.py — GCP error messages
• github/actions_deployer.py — GitHub Actions deployer (PyGithub + GitPython SDK)
• github/constants.py — GitHub error messages

Refactored Providers

• providers/storage_s3.py — Now uses cloud.aws.services.s3.S3
• providers/storage_gcs.py — Now uses cloud.gcp.services.gcs.GCS

Settings & Logging

• settings.py — Added TEMPLATE_REPO, TEMPLATE_BRANCH, TEMPLATE_CLOUD_DIR, DOTFLOW_OUTPUT_PATH env var
• logging.py — Removed mkdir and file handler (fixes Lambda read-only filesystem)

Documentation

• docs/nav/cloud/ — 34 platform pages (Docker, AWS Lambda/ECS variants, GCP Cloud Run, Kubernetes, GitHub Actions, Azure, Heroku, Fly.io, Railway, Render, DigitalOcean, and more)
• docs/nav/cloud/index.md — Overview with cards grouped by provider
• docs/nav/concepts/concept-providers.md — Updated with Tracer, Metrics, Sentry, pip extras table
• docs/nav/how-to/index.md — Removed Providers card (moved to Integrations)
• README.md — Revamped with sales pitch, cloud deploy table, observability, features in expand, documentation links
• docs/index.md — Replaced gif with png

Tests

• tests/cloud/aws/ — 18 tests (LambdaDeployer, LambdaApiDeployer, LambdaS3Deployer, LambdaSQSDeployer, ECSDeployer, S3 service)
• tests/cloud/gcp/ — 4 tests (GCS service)
• tests/cloud/github/ — 6 tests (ActionsDeployer)
• tests/providers/ — Updated StorageS3 and StorageGCS tests for new architecture

Optional Dependencies

• deploy-aws — boto3
• deploy-gcp — google-cloud-run, google-cloud-build, google-cloud-artifact-registry, google-cloud-service-usage, google-cloud-storage, google-auth
• deploy-github — PyGithub, GitPython

Motivation and Context

Enables users to generate infrastructure files and deploy dotflow pipelines to any major cloud provider from the CLI. Templates are hosted in dotflow-io/template and fetched at runtime.

The cloud module follows a clean architecture with ABCs for all services (Registry, RoleManager, LogManager, ObjectStorage, Queue, EventScheduler, Deployer), allowing new cloud providers to be added by implementing the interfaces.

Closes #173

Types of changes

• New feature (change which adds functionality)
• Documentation

Checklist

• I have performed a self-review of my own code
• I have added tests that prove my fix is effective or that my feature works
• I have updated the CHANGELOG
• I have updated the documentation accordingly