tests: make test_multi_attach more stable (#9202)

## Problem

`test_multi_attach` is sometimes failing with `invalid compute status
for configuration request: Configuration`. This is likely a result of
the test attempting to reconfigure the compute at the same time as the
storage controller is doing so.

This test was originally written before the storage controller existed,
and is not expecting anything else to be reconfiguring computes at the
same time.

## Summary of changes

- Configure the tenant into scheduling policy `Stop` in the storage
controller at the start of the test, so that it won't try to do anything
to the tenant while the test is running.

test_runner/regress/test_pageserver_generations.py

@@ -549,6 +549,14 @@ def test_multi_attach(
     tenant_id = env.initial_tenant
     timeline_id = env.initial_timeline
 
+    # Instruct the storage controller to not interfere with our low level configuration
+    # of the pageserver's attachment states.  Otherwise when it sees nodes go offline+return,
+    # it would send its own requests that would conflict with the test's.
+    env.storage_controller.tenant_policy_update(tenant_id, {"scheduling": "Stop"})
+    env.storage_controller.allowed_errors.extend(
+        [".*Scheduling is disabled by policy Stop.*", ".*Skipping reconcile for policy Stop.*"]
+    )
+
     # Initially, the tenant will be attached to the first pageserver (first is default in our test harness)
     wait_until(10, 0.2, lambda: assert_tenant_state(http_clients[0], tenant_id, "Active"))
     _detail = http_clients[0].timeline_detail(tenant_id, timeline_id)