fix(nuclei): nuclei can crash in some conditions

[ocervell]

With some specific templates and conditions, nuclei can crash when -hang-monitor is enabled because it detects 630 hanging go-routines, that are waiting for a server to respond, and ends up killing the process.

[bpce-it-dis-soc-sof]

We had the same issue, and in our case the problem was:

• Running OAST templates in a constrained environment (no OAST server reachable) --> some templates took 48s to run
• When running massively on a huge number of hosts, the hang-monitor timeout (30s) was lower than the 48s, and multiplied by > 630 targets was enough for nuclei to think that it's hanging
• Nuclei ends up killing itself

Solution: add the -ni flag (#1095) and pass it to our scan (or add it to a profile).

Maybe we could consider an internal profile as well, that would skip:

• CVE checks from cve.circl.lu
• OAST templates (-ni)
• Any other remote calls that could happen in Secator

[ocervell]

Thanks for digging, but I still got the issue even when excluding OAST templates with -ni. Most of the issues seem to come from RPC templates, I still need to figure out which ones.