fix(appolly): preserve other selector criteria with pid matches

[MrAlias]

Summary

• keep PID selectors as a narrowing criterion instead of letting them bypass the rest of the selector
• add regression coverage to ensure PID matches still require the other configured criteria to match

Why

Process matching returned early when a selector included target_pids, which let a PID hit short-circuit the rest of the selector criteria. That made mixed selectors behave too broadly whenever a matching PID was present.

This behavior also conflicted with the documented selector semantics. The generated OBI config reference describes target_pids as additive: the PID must match "in addition to any path/port criteria". The published OBI service discovery docs also state that when multiple selectors are configured in the same instrument entry, the process must match all selector properties.

Impact

This keeps selector evaluation consistent with the documented configuration model: a configured PID list can narrow matching, but it no longer overrides other path, language, command-argument, container, or metadata criteria.

Validation

• go test ./pkg/appolly/discover

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 69.23%. Comparing base (81871be) to head (31d19f7).
⚠️ Report is 13 commits behind head on main.

❗ There is a different number of reports uploaded between BASE (81871be) and HEAD (31d19f7). Click for more details.

HEAD has 8 uploads less than BASE

Flag	BASE (81871be)	HEAD (31d19f7)
integration-test-vm-x86_64-6.10.6	4	0
integration-test-vm-x86_64-5.15.152	4	0

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1911      +/-   ##
==========================================
- Coverage   78.23%   69.23%   -9.00%     
==========================================
  Files         278      279       +1     
  Lines       34240    34283      +43     
==========================================
- Hits        26786    23735    -3051     
- Misses       6205     9279    +3074     
- Partials     1249     1269      +20     

Flag	Coverage Δ
integration-test	56.09% <33.33%> (+0.70%)	⬆️
integration-test-arm	28.51% <33.33%> (-1.66%)	⬇️
integration-test-vm-x86_64-5.15.152	?
integration-test-vm-x86_64-6.10.6	?
k8s-integration-test	42.34% <33.33%> (-0.12%)	⬇️
oats-test	38.84% <33.33%> (+1.18%)	⬆️
unittests	58.68% <100.00%> (+0.36%)	⬆️

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

☔ 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.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[Copilot]

Pull request overview

Adjusts AppO11y process discovery selector semantics so target_pids narrows matches rather than short-circuiting other selector criteria, aligning runtime behavior with documented configuration expectations.

Changes:

• Update Matcher.matchProcess to treat PID lists as a required filter (fail fast on PID mismatch) while still evaluating other selector criteria.
• Add a regression test ensuring PID matches do not bypass other configured criteria (e.g., exe_path).

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated no comments.

File	Description
pkg/appolly/discover/matcher.go	Changes PID handling from early-return “match-by-PID only” to “PID must match, then evaluate remaining criteria”.
pkg/appolly/discover/matcher_test.go	Adds regression coverage verifying PID hits still require other selector properties (e.g., path) to match.

[damemi]

sorry about the confusion, I looked back and we did this purposefully because PID seemed pretty explicit: #1321 (comment) but I agree with the revert to match the documentation and other behavior