Verify op intent changes are always shared with the DSS

[BenjaminPelletier]

Is your feature request related to a problem? Please describe.
Despite our current checks and #1377 and #1378, a USS could potentially change the details of its operational intent without sharing that change with the DSS, and therefore be in violation of USS0105,1 and/or USS0005 because the op intent change was not made discoverable using the DSS, but uss_qualifier will not detect any failure.

As a specific example, a USS could provide op intent information where the reference time extent is consistently 1pm-2pm, but the details extent changes to a different 30-minute interval between 1pm and 2pm every time the USS provides op intent details (via query or notification). I do not think uss_qualifier would detect this problem currently unless one of the 30-minute intervals failed to cover the time range in which the flight planning was requested.

Describe the solution you'd like
uss_qualifier should, whenever convenient, check if the details of an op intent have changed without that op intent's version/OVN changing also. In other words, if the op intent's version/OVN are the same as a previous observation, then verify that the full op intent information exactly¹ matches the previously-obtained information for this version of the op intent in the cache. This could perhaps be accomplished by retaining a global/test-run-scope cache of op intent details indexed by op intent ID + version, then verifying identical content if the same op intent ID + version is ever encountered (via separate query to USS or notification from USS) again during the test.

Note that op intent version and OVN should change in lockstep, but a failure for that to happen (OVN changes when version didn't or version changes when OVN didn't) is a DSS failure rather than a USS failure (or #1378 if the USS is transmitting DSS information incorrectly). Since this Issue is regarding USS behavior, the solution to this Issue specifically should check whether version/OVN are the same merely to determine whether to perform the full op intent information comparison, and not try to detect a version change without OVN change or OVN change without version change.

Describe alternatives you've considered
A global cache of op intent details could perhaps get large over the course of an entire test run; it would probably be nicer to scope the cache to a single test scenario (since op intents are unlikely to intentionally persist between scenarios).

¹But we should be tolerant of numerical precision and formatting differences expressing fundamentally the same value.