chore(tests) split out plugin map test

[rainest]

Separate the plugin arbitrary map default test from the main default test. The new test only checks the arbitrary map field.

Changes to the test plugin in upstream Kong break the blanket diff used in the main test, and would require a version split without testing only the relevant field.

The issue appeared in https://github.com/Kong/go-kong/actions/runs/4718806033/jobs/8369120871?pr=315

 === RUN   TestFillPluginDefaults/nested_config_with_arbitrary_map_field
    plugin_service_test.go:706:   &kong.Plugin{
          	... // 5 identical fields
          	Service:  nil,
          	Consumer: nil,
          	Config: kong.Configuration{
          		"content_type":         string("application/json"),
          		"custom_fields_by_lua": map[string]any{"foo": string("bar")},
        - 		"flush_timeout":        nil,
        + 		"flush_timeout":        float64(2),
          		"headers":              nil,
          		"http_endpoint":        nil,
          		"keepalive":            float64(60000),
          		"method":               string("POST"),
        - 		"queue": map[string]any{
        - 			"initial_retry_delay":  float64(0.01),
        - 			"max_batch_size":       float64(1),
        - 			"max_bytes":            nil,
        - 			"max_coalescing_delay": float64(1),
        - 			"max_entries":          float64(10000),
        - 			"max_retry_delay":      float64(60),
        - 			"max_retry_time":       float64(60),
        - 			"name":                 nil,
        - 		},
        - 		"queue_size":  nil,
        + 		"queue_size":  float64(1),
        - 		"retry_count": nil,
        + 		"retry_count": float64(10),
          		"timeout":     float64(10000),
          	},
          	Enabled: &false,
          	RunOn:   nil,
          	... // 3 identical fields
          }
--- FAIL: TestFillPluginDefaults (0.04s)

These fields were added upstream for 3.3: Kong/kong@9df893f

[codecov-commenter]

Codecov Report

Patch and project coverage have no change.

Comparison is base (9458342) 52.73% compared to head (f7d9ae1) 52.73%.

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #316   +/-   ##
=======================================
  Coverage   52.73%   52.73%           
=======================================
  Files          68       68           
  Lines        5012     5012           
=======================================
  Hits         2643     2643           
  Misses       1798     1798           
  Partials      571      571           

Flag	Coverage Δ
2.1.4	35.87% <ø> (ø)
2.2.1.3	46.52% <ø> (ø)
2.2.2	16.75% <ø> (-19.12%)	⬇️
2.3.3	35.55% <ø> (-0.48%)	⬇️
2.3.3.4	47.20% <ø> (ø)
2.4.1	36.11% <ø> (ø)
2.4.1.3	47.28% <ø> (+0.07%)	⬆️
2.5.1.2	47.28% <ø> (+0.07%)	⬆️
2.5.2	36.11% <ø> (ø)
2.6.1	36.11% <ø> (ø)
2.6.1.0	47.28% <ø> (+0.07%)	⬆️
2.7.2	36.11% <ø> (ø)
2.7.2.0	48.84% <ø> (+0.07%)	⬆️
2.8.3	36.11% <ø> (ø)
2.8.4.0	48.84% <ø> (+0.07%)	⬆️
3.0.2	35.55% <ø> (ø)
3.0.2.0	49.30% <ø> (+0.07%)	⬆️
3.1.1	37.07% <ø> (ø)
3.1.1.3	50.91% <ø> (+0.07%)	⬆️
3.2.2	37.13% <ø> (ø)
3.2.2.1	50.91% <ø> (+0.07%)	⬆️
community	37.68% <ø> (ø)
enterprise	51.47% <ø> (+0.07%)	⬆️
enterprise-nightly	50.91% <ø> (+0.07%)	⬆️
integration	52.73% <ø> (ø)
nightly	37.13% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

[pmalek]

I've pushed the change to use require and assert based on correct (subtest's) *testing.T.

[czeslavo]

I allow myself to merge this - it unblocks PRs to main.