[EDR Workflows][E2E] Increase the timeout of agent check in (elastic#…

…168438)

This pull request extends the agent fleet check timeout from 2 minutes
to 4 minutes. We've identified a number of unreliable tests that fail
during the `beforeAll` stage while executing the `createEndpointHost`
task. The following logs appear before the timeout:

```
info Enrolling Elastic Agent with Fleet
  | Installing service....... DONE
  | Starting service... DONE
  | Enrolling Elastic Agent with Fleet..........Successfully enrolled the Elastic Agent.
  | Elastic Agent has been successfully installed.
  | info Waiting for Agent to check-in with Fleet
```

The error message we encounter is `> Timed out waiting for host
[test-host-4981] to appear in Fleet.`

It appears that all the preceding steps are successful, and only the
final one fails due to either the agent not checking in with the fleet
for 2 minutes or the agent being unhealthy for two minutes. Since I
haven't been able to replicate this behavior locally, and there isn't a
way to inspect what's happening on the agent, I believe the best course
of action at this point is to extend the timeout and monitor the
results.

Reports of this error:
closes elastic#168427
closes elastic#168394
closes elastic#168393
closes elastic#168390
closes elastic#168363
closes elastic#168362
closes elastic#168361
closes elastic#168360
closes elastic#168359

Affected CI runs:
https://buildkite.com/elastic/kibana-on-merge/builds/36483
https://buildkite.com/elastic/kibana-on-merge/builds/36497
https://buildkite.com/elastic/kibana-on-merge/builds/36501
https://buildkite.com/elastic/kibana-on-merge/builds/36526

Another time out happens from time to time when previously set 10
minutes timeout on `createEndpointHost` task is not enough to set up the
environment. Its portrayed below, timeout happens during agent setup
```
  | default: Running: inline script
  | default: Reading package lists...
  | default: Building dependency tree...
  | default: Reading state information...
  | default: Suggested packages:
  | default:   zip
  | default: The following NEW packages will be installed:
  | default:   unzip
  | default: 0 upgraded, 1 newly installed, 0 to remove and 0 not upgraded.
  | default: Need to get 174 kB of archives.
  | default: After this operation, 385 kB of additional disk space will be used.
  | default: Get:1 http://archive.ubuntu.com/ubuntu jammy-updates/main amd64 unzip amd64 6.0-26ubuntu3.1 [174 kB]
  | default: dpkg-preconfigure: unable to re-open stdin: No such file or directory
  | default: Fetched 174 kB in 1s (210 kB/s)
  | default: Selecting previously unselected package unzip.
  | (Reading database ... 63961 files and directories currently installed.)
  | default: Preparing to unpack .../unzip_6.0-26ubuntu3.1_amd64.deb ...
  | default: Unpacking unzip (6.0-26ubuntu3.1) ...
  | default: Setting up unzip (6.0-26ubuntu3.1) ...
  | default: Processing triggers for man-db (2.10.2-1) ...
  |  
  | CypressError: `cy.task('createEndpointHost')` timed out after waiting `600000ms`.
```

x-pack/plugins/security_solution/public/management/cypress/tasks/create_endpoint_host.ts

@@ -17,6 +17,6 @@ export const createEndpointHost = (
     {
       agentPolicyId,
     },
-    { timeout: timeout ?? 600000 }
+    { timeout: timeout ?? 900000 } // 15 minutes, since setup can take 10 minutes and more. Task will time out if is not resolved within this time.
   );
 };

x-pack/plugins/security_solution/scripts/endpoint/common/endpoint_host_services.ts

@@ -335,7 +335,7 @@ const enrollHostWithFleet = async ({
     ]);
   }
   log.info(`Waiting for Agent to check-in with Fleet`);
-  const agent = await waitForHostToEnroll(kbnClient, vmName, 120000);
+  const agent = await waitForHostToEnroll(kbnClient, vmName, 240000);
 
   return {
     agentId: agent.id,