[platform] Improve CSV mapper behavior to allow correct count and failure tracking (#6844)

[richard-julien]

See #6844 and OpenCTI-Platform/client-python#629

[codecov]

Codecov Report

Attention: Patch coverage is 33.33333% with 50 lines in your changes are missing coverage. Please review.

Project coverage is 67.67%. Comparing base (504101c) to head (b17308d).
Report is 3 commits behind head on master.

Files	Patch %	Lines
...hql/src/connector/importCsv/importCsv-connector.ts	0.00%	28 Missing ⚠️
...latform/opencti-graphql/src/types/stix-common.d.ts	0.00%	18 Missing ⚠️
...-platform/opencti-graphql/src/parser/csv-parser.ts	70.00%	3 Missing ⚠️
...platform/opencti-graphql/src/parser/csv-bundler.ts	94.44%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #6862      +/-   ##
==========================================
+ Coverage   67.48%   67.67%   +0.19%     
==========================================
  Files         561      561              
  Lines       68734    68741       +7     
  Branches     5774     5990     +216     
==========================================
+ Hits        46384    46522     +138     
+ Misses      22350    22219     -131     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[lndrtrbn]

I don't know how to test a complex bundle with errors in it to test the use case but locally tested normal csv mappers and import data works correctly

[SouadHadjiat]

import_bundle_from_json returns a list of elements, not a number, and imported_items is initialized at 0. is it a list or a number ?

[richard-julien]

Its a list, the initialization to 0 must be an empty array. Thanks

[labo-flg]

I pushed the rebase to master ; conflict in requirements.txt.

[labo-flg]

I pushed the rebase to master ; conflict in requirements.txt.

CI is red, we probably need to rebase also the corresponding branch in python-client