fix: stop download on action timeout, don't drop action on resend

[de-sh]

Closes #

Changes

1. Timeout downloads at expected timeout of action on bridge
2. Don't clear action on unsuccessful re-send even if with same id

Why?

Downloader as of right now continues to run despite the action having timedout on bridge, leading to further action failures as download actions clog the channel into the downloader

Trials Performed

1. Create http server accessible over toxiproxy.

echo '#!/bin.sh' > test.sh
echo 'echo "Hello, World!"' >> test.sh
python3 -m http.server 8000

curl --request POST \
  --url http://localhost:8474/proxies \
  --header 'Content-Type: application/json' \
  --data '{
    "name": "script-server",
    "listen": "0.0.0.0:8001",
    "upstream": "localhost:8000",
    "enabled": true
}'

3. Start uplink without auth.

# auth.json
{
  "project_id": "demo",
  "broker": "localhost",
  "port": 1883,
  "device_id":"1"
}
# config.toml
[downloader]
actions = [{ name = "send_script" }]
path = "/tmp/downloads"

4. Send action to uplink that points to http sever.

mosquitto_pub -t /tenants/demo/devices/1/actions -m '{ "action_id":
 "1", "kind": "process", "name": "send_script", "payload": "{\"file_n
ame\":\"test.sh\",\"content-length\":31,\"id\":\"123\",\"url\":\"http
://0.0.0.0:8001/test.sh\"}" }'

5. Ensure at normal speeds this doesn't trigger failure.
6. Repeat step 3 with extremely slow speed access, see if this triggers failure and download continues to run despite action timeout in bridge.

  2023-07-25T16:18:02.531840Z ERROR uplink::collector::downloader: Last download has timedout

  2023-07-25T16:18:02.532611Z ERROR uplink::base::bridge: Timeout waiting for action response. Action ID = 1