Automation Update

[lkingland]

Our primary repository's automation is flaky, has several actions with overlapping functionality, and local make targets have diverged from those run in CI. In general we should probably:

• Refactor integration with codecov (it appears the integration is probabalistic and often fails to report a large chunk of the coverage, resulting in manual overrides)
• Remove Functions-specific actions which are now covered by the Knative common tooling.
• Update the Makefile targets to run the internals of these tests as closely as possible locally (i.e. make test should be effectively equialent to the GitHub Action, ideally running the same script if possible. This would, for example run linters with the same options locally as in CI).
• Transition to using the shared Knatve quick-start for allocating the cluster, in hopes it is less flaky (cluster allocation often fails with timeouts due to apparently having allocated resources of a potato).
• Update the build/test flags to be test with or without integration tests, rather than the current configuration of disjunct sets.
• Review our E2E tests to identify areas of possible optimization
• Investigate if we can disable the Prow bot's auto-assign of reviewers