feat: prefetch and cache media resources

[Teagan42]

What

• compute dense and sparse search embeddings concurrently
• prefetch extra Qdrant results and cache media payloads, posters, and backgrounds
• test concurrency and cache reuse

Why

• speeds up search and subsequent resource lookups by parallelizing work and avoiding duplicate queries

Affects

• mcp_plex/server.py
• tests/test_server.py

Testing

• uv run ruff check .
• uv run pytest

Documentation

• n/a

---

https://chatgpt.com/codex/tasks/task_e_68b3e4d992dc8328b28e420e21bd16e7

[github-actions]

Coverage Report

File	Stmts	Miss	Cover	Missing
mcp_plex
loader.py	179	10	94%	32–34, 296, 299–302, 523, 529
server.py	136	25	82%	36, 39, 57–58, 92–102, 264–268, 286–290
TOTAL	417	35	92%

Tests	Skipped	Failures	Errors	Time
11	0 💤	0 ❌	0 🔥	6.344s ⏱️