Add pod logging (dagster-io#26196)

## Summary & Motivation
This is a follow up of dagster-io#22784
Such logging is still necessary to have even after
`delete_failed_k8s_jobs` implementation, because such option isn't
feasible in production or/and highly dynamic environments.

Addressed previous comments of @gibsondan 

## How I Tested These Changes
Tested locally by inducing an error. Logging was successful as expected.

## Changelog

Output k8s pod logs when pods fail.

python_modules/libraries/dagster-k8s/dagster_k8s/ops/k8s_job_op.py

@@ -419,6 +419,20 @@ def execute_k8s_job(
             num_pods_to_wait_for=num_pods_to_wait_for,
         )
     except (DagsterExecutionInterruptedError, Exception) as e:
+        try:
+            pods = api_client.get_pod_names_in_job(job_name=job_name, namespace=namespace)
+            pod_debug_info = "\n\n".join(
+                [api_client.get_pod_debug_info(pod_name, namespace) for pod_name in pods]
+            )
+        except Exception:
+            context.log.exception(
+                f"Error trying to get pod debug information for failed k8s job {job_name}"
+            )
+        else:
+            context.log.error(
+                f"Debug information for failed k8s job {job_name}:\n\n{pod_debug_info}"
+            )
+
         if delete_failed_k8s_jobs:
             context.log.info(
                 f"Deleting Kubernetes job {job_name} in namespace {namespace} due to exception"