Fix: header behaviour in openapi2kong

[harshadixit12]

Second attempt at #285, which was reverted due to failing tests.

Fixes:
Kong/deck#1856
Kong/deck#1838

[codecov-commenter]

Codecov Report

❌ Patch coverage is 90.00000% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 69.82%. Comparing base (a6a5619) to head (a15fc97).

Files with missing lines	Patch %	Lines
openapi2kong/openapi2kong.go	90.00%	1 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #289      +/-   ##
==========================================
+ Coverage   69.75%   69.82%   +0.07%     
==========================================
  Files          24       24              
  Lines        2738     2751      +13     
==========================================
+ Hits         1910     1921      +11     
- Misses        652      653       +1     
- Partials      176      177       +1     

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[harshadixit12]

I've added a commit to fix duplicate plugin IDs in older tests.
Also - in 6463172#diff-c655e44d9e55d9766b44a0cff7734f9e880cc5185762cc3a51f462deea661492 we are fixing a test - because based on implementation when there are >1 headers - say one required (m variants in enum), and one optional, we are creating m routes - no fallback.

[mheap]

@harshadixit12 Do you think the implemented behaviour is correct here? Should we have a fallback?

[harshadixit12]

@harshadixit12 Do you think the implemented behaviour is correct here? Should we have a fallback?

Since this header behaviour is limited to required headers - technically this may be ok. However, I imagine APIs don't start out with required headers - for example, a version header would only become required after there is > 1 version of API live.
Let me add a fallback.