[DOCS] Clarify ML job closure prerequisites (#49265)

Author: lcawl

docs/reference/ml/anomaly-detection/apis/close-job.asciidoc

@@ -27,6 +27,8 @@ operations, but you can still explore and navigate results.
 * If the {es} {security-features} are enabled, you must have `manage_ml` or
 `manage` cluster privileges to use this API. See
 <<security-privileges>>.
+* Before you can close an {anomaly-job}, you must stop its {dfeed}. See
+<<ml-stop-datafeed>>.
 
 [[ml-close-job-desc]]
 ==== {api-description-title}

docs/reference/ml/anomaly-detection/apis/delete-job.asciidoc

@@ -18,6 +18,9 @@ Deletes an existing {anomaly-job}.
 
 * If {es} {security-features} are enabled, you must have `manage_ml` or `manage`
 cluster privileges to use this API. See <<security-privileges>>.
+* Before you can delete a job, you must delete the {dfeeds} that are associated
+with it. See <<ml-delete-datafeed>>.
+* Before you can delete a job, you must close it (unless you specify the `force` parameter). See <<ml-close-job>>.
 
 [[ml-delete-job-desc]]
 ==== {api-description-title}
@@ -29,10 +32,6 @@ delete the job directly from the `.ml-*` indices using the {es} delete document
 API. When {es} {security-features} are enabled, make sure no `write` privileges
 are granted to anyone over the `.ml-*` indices.
 
-Before you can delete a job, you must delete the {dfeeds} that are associated
-with it. See <<ml-delete-datafeed,Delete {dfeeds-cap}>>. Unless the `force`
-parameter is used the job must be closed before it can be deleted.
-
 It is not currently possible to delete multiple jobs using wildcards or a comma
 separated list.