Fix task/logmon leak after crash

Some of the context uses in TR hooks are useless (Killed during Stop
never seems meaningful).

None of the hooks are interruptable for graceful shutdown which is
unfortunate and probably needs fixing.

Backward compat note about task config validation

artifact stanza documentation

Note rkt driver needs Nomad client run as root

Logs view - support widescreen format

The rkt driver requires that the Nomad agent to be running as the root user

drivers/shared/executor: fix strings.Replace call

nomad: compare current eval when setting WaitIndex

Made small adjustment to make it clear that 0.8.7 would require the legacy syntax and that the deprecation notice was more about the legacy syntax becoming unsupported at some point after v0.9.0

Add missing period to sentence.

To pick up https://github.com/hashicorp/hcl2/pull/90

Co-Authored-By: schmichael <michael.schurter@gmail.com>

Co-Authored-By: schmichael <michael.schurter@gmail.com>

Builds upon earlier commit that cleans up restored handles of terminal
allocs by also emitting terminated events and calling exited hooks when
appropriate.

Co-Authored-By: schmichael <michael.schurter@gmail.com>

Logged once per terminal task on agent restart.

The test is sadly quite complicated and peeks into things (logmon's
reattach config) AR doesn't normally have access to.

However, I couldn't find another way of asserting logmon got cleaned up
without resorting to smaller unit tests. Smaller unit tests risk
re-implementing dependencies in an unrealistic way, so I opted for an
ugly integration test.

s/mananger/manager/g

Consider currently dequeued Evaluation's ModifyIndex when determining
its WaitIndex. Normally the Evaluation itself would already be in the
state store snapshot used to determine the WaitIndex. However, since the FSM
applies Raft messages to the state store concurrently with Dequeueing,
it's possible the currently dequeued Evaluation won't yet exist in the
state store snapshot used by JobsForEval.

This can be solved by always considering the current eval's modify index
and using it if it is greater than all of the evals returned by the
state store.

This commit is a significant change. TR.Run is now always executed, even
for terminal allocations. This was changed to allow TR.Run to cleanup
(run stop hooks) if a handle was recovered.

This is intended to handle the case of Nomad receiving a
DesiredStatus=Stop allocation update, persisting it, but crashing before
stopping AR/TR.

The commit also renames task runner hook data as it was very easy to
accidently set state on Requests instead of Responses using the old
field names.

strings.Replace call with n=0 argument makes no sense
as it will do nothing. Probably -1 is intended.
Signed-off-by: Iskander Sharipov <quasilyte@gmail.com>

drivers/docker: rename logging `type` to `driver`

artifact docs: modified the leading example and removed a treacherous (but potentially correct) example

Docker uses the term logging `driver` in its public documentations: in
`docker` daemon config[1], `docker run` arguments [2] and in docker compose file[3].
Interestingly, docker used `type` in its API [4] instead of everywhere
else.

It's unfortunate that Nomad used `type` modeling after the Docker API
rather than the user facing documents.  Nomad using `type` feels very
non-user friendly as it's disconnected from how Docker markets the flag
and shows internal representation instead.

Here, we rectify the situation by introducing `driver` field and
prefering it over `type` in logging.

[1] https://docs.docker.com/config/containers/logging/configure/
[2] https://docs.docker.com/engine/reference/run/#logging-drivers---log-driver
[3] https://docs.docker.com/compose/compose-file/#logging
[4] https://docs.docker.com/engine/api/v1.39/#operation/ContainerCreate