netmap: packet stalls (6.0.x backport) #5863 v1

[bmeeks8]

Make sure these boxes are signed before submitting your Pull Request -- thank you.

• I have read the contributing guide lines at https://suricata.readthedocs.io/en/latest/devguide/codebase/contributing/contribution-process.html
• I have signed the Open Information Security Foundation contribution agreement at https://suricata.io/about/contribution-agreement/
• I have updated the user guide (in doc/userguide/) to reflect the changes made (if applicable)

Link to redmine ticket: https://redmine.openinfosecfoundation.org/issues/5863

Describe changes:

• Fix packet processing stall under high load when using netmap in IPS mode (see Redmine #5863 and related issue #5744).
• Detect and generate FatalError() exit for the rare case when hardware NICs expose unmatched RX/TX queue counts (for example, exposing more TX queues than RX queues). This is a rare possibility, but if not detected and handled it would result in some netmap traffic bypassing Suricata since the code previously assumed NIC queue counts were always symmetrical.
• Fix instance of missing unlock calls for the netmap device list when exiting a function due to an error condition.
• Clean up and/or add some comments to more accurately reflect what's happening in the new netmap code.

[bmeeks8]

Replaced by #8551.