ci(deploy): switch deploy trigger from push-to-main to manual workflow_dispatch (#39)

Remove the push-to-main trigger from .github/workflows/deploy.yaml so merges
no longer deploy automatically; deploys now run only via the Actions "Run
workflow" button (workflow_dispatch). No job content, permissions, or
reusable-workflow inputs changed. README and the auth deploy runbook updated
to match.

.github/workflows/deploy.yaml

@@ -1,9 +1,7 @@
 name: deploy
 
 on:
-  push:
+  workflow_dispatch:
-    branches:
-      - main
 
 permissions:
   contents: read

README.md

@@ -63,7 +63,8 @@ npx cdk deploy sh-mcp-auth \
 ```
 
 CI runs `npx cdk synth` at the repo root on every PR (the `ci-typescript-cdk` reusable workflow with
-`run-cdk-synth: true`); merges to `main` deploy via the `cd-cdk` reusable workflow over the
+`run-cdk-synth: true`); deploys are triggered manually (Actions → deploy → Run workflow on
+`main`, i.e. `workflow_dispatch`) via the `cd-cdk` reusable workflow over the
 per-repo `githubdeploy-sh-mcp` OIDC role. **Nothing is deployed to AWS yet** — see status above.
 
 ## Auth substrate (Phase 2a — `infra/` + `auth/`)

docs/runbook-auth-deploy.md

@@ -11,11 +11,11 @@ validated."
 - **Toolchain:** aws-cdk `2.1128.1` (CLI, root devDep), aws-cdk-lib `2.260.0`, constructs `10.6.0`.
 
 > **CD is already wired and currently red.** `.github/workflows/deploy.yaml`
-> (push-to-`main` → org reusable `cd-cdk.yaml@main`, OIDC) has failed on every
+> (manual `workflow_dispatch` → org reusable `cd-cdk.yaml@main`, OIDC) has failed
-> merge so far (~4s) because the `githubdeploy-sh-mcp` deploy role does not exist
+> on every run so far (~4s) because the `githubdeploy-sh-mcp` deploy role does not
-> yet (Prereq 4). Once the prerequisites below are in place, CD deploys on the
+> exist yet (Prereq 4). Once the prerequisites below are in place, CD is triggered
-> next push to `main`; the first deploy is done **manually** (Step 2) so a human
+> manually via the Actions "Run workflow" button on `main`; the first deploy is
-> watches the initial resource creation.
+> done **manually** (Step 2) so a human watches the initial resource creation.
 
 ---
 
@@ -162,12 +162,13 @@ grants). Note the stack outputs (exported for 2b cross-stack import):
 
 ## Step 3 — hand off to CD
 
-After the manual first deploy succeeds and Prereq 4 is in place, every push to
+After the manual first deploy succeeds and Prereq 4 is in place, deploys are
-`main` deploys via `.github/workflows/deploy.yaml`. Re-run the last failed deploy
+triggered manually via `.github/workflows/deploy.yaml` (Actions → deploy → Run
+workflow on `main`, i.e. `workflow_dispatch`). Re-run the last failed deploy
 to confirm it now goes green:
 
 ```bash
-gh workflow run deploy.yaml --ref main      # or push any commit to main
+gh workflow run deploy.yaml --ref main
 gh run watch $(gh run list --workflow=deploy.yaml --limit 1 --json databaseId --jq '.[0].databaseId')
 ```
 
@@ -221,7 +222,8 @@ aws dynamodb delete-item --table-name sh-mcp-deny-list --region us-east-1 \
   a later redeploy re-adopts them). Delete the tables manually only if you intend
   to lose revocation/freshness state.
 - A bad deploy rolls back automatically (CloudFormation). To revert code, revert
-  the commit on `main`; CD redeploys the prior template.
+  the commit on `main`, then run the deploy workflow manually (Actions → deploy →
+  Run workflow) to redeploy the prior template.
 - **0a spike teardown is the LAST step, not part of this deploy.** The live 0a
   spike kit (Cognito pool `us-east-1_GsDbGe0pa`, probe Lambda/API, the SF
   `sh_mcp_0a` objects) stays up until `sh-mcp-auth` is deployed **and validated**