Upgrade the operator to use Operator SDK v1.33.0 (#182)

* Move controller package inside internal directory

Based on the go/v4 project structure, the following changed:
- Pakcage `controllers` is now named `controller`
- Package `controller` now lives inside new `internal` directory

* Move main.go in cmd directory

Based on the new go/v4 project structure, `main.go` now lives in the `cmd` directory.

* Change package import in main.go

* Update go mod dependencies

Update the dependencies based on the versions obtained by creating a new operator project using `kubebuilder init --domain onepassword.com --plugins=go/v4`.

This is based on the migration steps provided to go from go/v3 to go/v4 (https://book.kubebuilder.io/migration/migration_guide_gov3_to_gov4)

* Update vendor

* Adjust code for breaking changes from pkg update

sigs.k8s.io/controller-runtime package had breaking changes from v0.14.5 to v0.16.3. This commit brings the changes needed to achieve the same things using the new functionality avaialble.

* Adjust paths to connect yaml files

Since `main.go` is now in `cmd` directory, the paths to the files for deploying Connect have to be adjusted based on the new location `main.go` is executed from.

* Update files based on new structure and scaffolding

These changes are made based on the new project structure and scaffolding obtained when using the new go/v4 project structure.

These were done based on the migration steps mentioned when migrating to go/v4 (https://book.kubebuilder.io/migration/migration_guide_gov3_to_gov4).

* Update config files

These updates are made based on the Kustomize v4 syntax.

This is part of the upgrate to go/v4 (https://book.kubebuilder.io/migration/migration_guide_gov3_to_gov4)

* Update dependencies and GO version

* Update vendor

* Update Kubernetes tools versions

* Update operator version in Makefile

Now the version in the Makefile matches the version of the operator

* Update Operator SDK version in version.go

* Adjust generated deepcopy

It seems that the +build tag is no longer needed based on the latest generated scaffolding, therefore it's removed.

* Update copyright year

* Bring back missing changes from migration

Some customization in Makefile was lost during the migration process. Specifically, the namespace customization for `make deploy` command.

Also, we push changes to kustomization.yaml for making the deploy process smoother.

* Add RBAC perms for coordination.k8s.io

It seems that with the latest changes to Kubernetes and Kustomize, we need to add additional RBAC to the service account used so that it can properly access the `leases` resource.

* Optimize Dockerfile

Dockerfile had a step for caching dependencies (go mod download). However, this is already done by the vendor directory, which we include. Therefore, this step can be removed to make the image build time faster.

vendor/go.uber.org/zap/config.go

@@ -95,6 +95,32 @@ type Config struct {
 
 // NewProductionEncoderConfig returns an opinionated EncoderConfig for
 // production environments.
+//
+// Messages encoded with this configuration will be JSON-formatted
+// and will have the following keys by default:
+//
+//   - "level": The logging level (e.g. "info", "error").
+//   - "ts": The current time in number of seconds since the Unix epoch.
+//   - "msg": The message passed to the log statement.
+//   - "caller": If available, a short path to the file and line number
+//     where the log statement was issued.
+//     The logger configuration determines whether this field is captured.
+//   - "stacktrace": If available, a stack trace from the line
+//     where the log statement was issued.
+//     The logger configuration determines whether this field is captured.
+//
+// By default, the following formats are used for different types:
+//
+//   - Time is formatted as floating-point number of seconds since the Unix
+//     epoch.
+//   - Duration is formatted as floating-point number of seconds.
+//
+// You may change these by setting the appropriate fields in the returned
+// object.
+// For example, use the following to change the time encoding format:
+//
+//	cfg := zap.NewProductionEncoderConfig()
+//	cfg.EncodeTime = zapcore.ISO8601TimeEncoder
 func NewProductionEncoderConfig() zapcore.EncoderConfig {
 	return zapcore.EncoderConfig{
 		TimeKey:        "ts",
@@ -112,11 +138,22 @@ func NewProductionEncoderConfig() zapcore.EncoderConfig {
 	}
 }
 
-// NewProductionConfig is a reasonable production logging configuration.
-// Logging is enabled at InfoLevel and above.
+// NewProductionConfig builds a reasonable default production logging
+// configuration.
+// Logging is enabled at InfoLevel and above, and uses a JSON encoder.
+// Logs are written to standard error.
+// Stacktraces are included on logs of ErrorLevel and above.
+// DPanicLevel logs will not panic, but will write a stacktrace.
 //
-// It uses a JSON encoder, writes to standard error, and enables sampling.
-// Stacktraces are automatically included on logs of ErrorLevel and above.
+// Sampling is enabled at 100:100 by default,
+// meaning that after the first 100 log entries
+// with the same level and message in the same second,
+// it will log every 100th entry
+// with the same level and message in the same second.
+// You may disable this behavior by setting Sampling to nil.
+//
+// See [NewProductionEncoderConfig] for information
+// on the default encoder configuration.
 func NewProductionConfig() Config {
 	return Config{
 		Level:       NewAtomicLevelAt(InfoLevel),
@@ -134,6 +171,32 @@ func NewProductionConfig() Config {
 
 // NewDevelopmentEncoderConfig returns an opinionated EncoderConfig for
 // development environments.
+//
+// Messages encoded with this configuration will use Zap's console encoder
+// intended to print human-readable output.
+// It will print log messages with the following information:
+//
+//   - The log level (e.g. "INFO", "ERROR").
+//   - The time in ISO8601 format (e.g. "2017-01-01T12:00:00Z").
+//   - The message passed to the log statement.
+//   - If available, a short path to the file and line number
+//     where the log statement was issued.
+//     The logger configuration determines whether this field is captured.
+//   - If available, a stacktrace from the line
+//     where the log statement was issued.
+//     The logger configuration determines whether this field is captured.
+//
+// By default, the following formats are used for different types:
+//
+//   - Time is formatted in ISO8601 format (e.g. "2017-01-01T12:00:00Z").
+//   - Duration is formatted as a string (e.g. "1.234s").
+//
+// You may change these by setting the appropriate fields in the returned
+// object.
+// For example, use the following to change the time encoding format:
+//
+//	cfg := zap.NewDevelopmentEncoderConfig()
+//	cfg.EncodeTime = zapcore.ISO8601TimeEncoder
 func NewDevelopmentEncoderConfig() zapcore.EncoderConfig {
 	return zapcore.EncoderConfig{
 		// Keys can be anything except the empty string.
@@ -152,12 +215,15 @@ func NewDevelopmentEncoderConfig() zapcore.EncoderConfig {
 	}
 }
 
-// NewDevelopmentConfig is a reasonable development logging configuration.
-// Logging is enabled at DebugLevel and above.
+// NewDevelopmentConfig builds a reasonable default development logging
+// configuration.
+// Logging is enabled at DebugLevel and above, and uses a console encoder.
+// Logs are written to standard error.
+// Stacktraces are included on logs of WarnLevel and above.
+// DPanicLevel logs will panic.
 //
-// It enables development mode (which makes DPanicLevel logs panic), uses a
-// console encoder, writes to standard error, and disables sampling.
-// Stacktraces are automatically included on logs of WarnLevel and above.
+// See [NewDevelopmentEncoderConfig] for information
+// on the default encoder configuration.
 func NewDevelopmentConfig() Config {
 	return Config{
 		Level:            NewAtomicLevelAt(DebugLevel),