Fuzz dnp3 fixes v1

[catenacyber]

Does not proceed final chunk if we got an error previously
Flips the direction for last chunk as usual

Link to redmine tickets:
https://redmine.openinfosecfoundation.org/issues/3772
https://redmine.openinfosecfoundation.org/issues/3773
https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=22832

Describe changes:

• Fixes fuzz target fuzz_applayerparserparse by not processing a last chunk if we had an error previously
• Improves DNP3 fuzzing by not checking CRC while fuzzing
• Improves DNP3 probing parser by fixing direction based on header field
• Fixes DNP3 response parsing, as reported in dnp3: added sanity checks to avoid crashes in a case of malformed data #5063

So, this is a mix of DNP3 and fuzzing, and can be split between different Pull Requests

Should I make a S-V test for the probing parser ?

#suricata-verify-pr:
#suricata-verify-repo:
#suricata-verify-branch:
#suricata-update-pr:
#suricata-update-repo:
#suricata-update-branch:
#libhtp-pr:
#libhtp-repo:
#libhtp-branch:

[catenacyber]

Follows #5084 by adding dnp3 fix

[jasonish]

Should I make a S-V test for the probing parser ?

Yes, please. DNP3 lacks test coverage on imperfect input I believe.

[catenacyber]

Yes, please. DNP3 lacks test coverage on imperfect input I believe.

OISF/suricata-verify#261

[victorjulien]

Merged in #5139, thanks!