Use LRUCache for Application Manifests Cache

[Warashi]

What this PR does / why we need it:

Use LRUCache for the application manifests cache instead of TTLCache to control memory consumption.

Investigation details
The provided piped CPU/memory profiles are below.

Memory Profile

CPU Profile

The CPU profiles shows garbage collection takes times, so there seems a lot of long-living objects.

The memory profiles show that the drift detector of Kubernetes makes many memory allocations.

The allocated objects in kubernetes.(*loader).loadHeadManifests are stored in the application manifests cache, and the cache key contains the git commit hash.

pipecd/pkg/app/piped/driftdetector/kubernetes/detector.go

Lines 285 to 290 in f5dcaa4

	manifests, err = loader.LoadManifests(ctx)
	if err != nil {
	err = fmt.Errorf("failed to load new manifests: %w", err)
	return nil, err
	}
	manifestCache.Put(headCommit.Hash, manifests)

This cache is memorystore.TTLCache, so the cache eviction occurs only in the TTL mechanisms.
When the commit hash is changed many times quickly, the cache grows larger and larger.

This PR changes the cache from TTLCache to LRUCache, so the cache size is now limited.

Which issue(s) this PR fixes:

N/A

Does this PR introduce a user-facing change?:

• How are users affected by this change:
• Is this breaking change:
• How to migrate (if breaking change):

[codecov]

Codecov Report

Attention: Patch coverage is 0% with 11 lines in your changes missing coverage. Please review.

Project coverage is 22.82%. Comparing base (9709438) to head (1ef8e83).
Report is 23 commits behind head on master.

Files	Patch %	Lines
pkg/app/piped/cmd/piped/piped.go	0.00%	11 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #5108      +/-   ##
==========================================
- Coverage   22.82%   22.82%   -0.01%     
==========================================
  Files         409      409              
  Lines       43338    43343       +5     
==========================================
  Hits         9891     9891              
- Misses      32668    32673       +5     
  Partials      779      779              

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

[ffjlabo]

LGTM 🙆‍♂️