ISSUE-1830: Adding changes for stopping the seceret and malware scans

[varunsharma0286]

Fixes #
https://github.com/deepfence/enterprise-roadmap/issues/1830
Changes proposed in this pull request:

• Adding support for stopping the secret scans
• Adding support for stopping the malware scans

[noboruma]

This MaxStops is a bit misleading because it only applies for 1 iteration: Let's imagine we cancel 30 scans, we will do: 5 cancel, next 30 seconds 5 more, next 30 seconds 5 more, without checking if the previous cancel were completed.

Now I think in practice we should be fine because we expect cancels to finish quickly and the request should not use extra resources to complete.. I would rather remove that MaxStops entirely, just to simplify our logic and not mislead us in the future

[varunsharma0286]

I believe we would need to put in a limit to how many we can process in a one go.
Maybe we can raise this to higher number (maybe 50)

[noboruma]

I would rather keep previous code here as this is not solving anything new from what I understand.
The allocation optimization looks fine:

statuses := make([]map[string]interface{}, len(statusBuff))

But the Posn introduction is not solving anything new, is it?

[varunsharma0286]

This is to maintain the ordering of the messages.
Right now we store conflated data in the map, and maps does not guarantee the ordering of the messages.
For instance, if we received the message in below order:
"IN_PROGRESS" -> "CANCELLED" status, we are not guaranteed to process the status in the same order.
Hence we might end up with a record in the neo4j which is in "IN_PROGRESS" status.