fix(pack): add & as escape chunk filename character

[fireairforce]

Summary

Should not bundle & into output chunk filename, it will cause cdn link failed(character translation)

ref pr: utooland/next.js#146

Test Plan

add snapshot test case

[gemini-code-assist]

Code Review

This pull request introduces filename escaping for chunks and assets to handle special characters such as '&', supported by a new snapshot test. The review feedback highlights that the escaping logic is inconsistently applied, specifically missing in default code paths where no filename template is provided, and notes a redundant escaping operation in one instance.

[github-actions]

📊 Performance Benchmark Report (with-antd)

Utoopack Performance Report

Report ID: utoopack_performance_report_20260507_064420
Generated: 2026-05-07 06:44:20
Trace File: trace_antd.json (0.5GB, 1.57M spans)
Test Project: examples/with-antd

---

Executive Summary

Metric	Value	Assessment
Total Wall Time	11,680.4 ms	Baseline
Total Thread Work (de-duped)	40,716.1 ms	Non-overlapping busy time
Effective Parallelism	3.5x	thread_work / wall_time
Working Threads	5	Threads with actual spans
Thread Utilization	69.7%	🆗 Average
Total Spans	1,565,971	All B/E + X events
Meaningful Spans (>= 10us)	333,249	(21.3% of total)
Tracing Noise (< 10us)	1,232,722	(78.7% of total)

---

Build Phase Timeline

Shows when each build phase is active and how much CPU it consumes.
Self-Time is the time spent exclusively in that phase (excluding children).

Phase	Spans	Inclusive (ms)	Self-Time (ms)	Wall Range (ms)
Resolve	74,669	7,150.7	5,747.8	8,925.1
Parse	9,399	3,318.2	2,845.6	11,548.3
Analyze	201,511	30,609.3	22,089.8	11,231.0
Chunk	14,103	1,877.9	1,738.2	8,931.8
Codegen	27,746	3,764.6	2,779.8	8,990.8
Emit	39	45.6	22.6	8,976.2
Other	5,782	1,918.3	1,479.3	11,680.4

---

Workload Distribution by Diagnostic Tier

Category	Spans	Inclusive (ms)	% Work	Self-Time (ms)	% Self
P0: Scheduling & Resolution	279,674	39,039.6	95.9%	28,697.8	70.5%
P1: I/O & Heavy Tasks	3,034	232.3	0.6%	209.4	0.5%
P2: Architecture (Locks/Memory)	0	0.0	0.0%	0.0	0.0%
P3: Asset Pipeline	49,597	8,844.2	21.7%	7,247.3	17.8%
P4: Bridge/Interop	0	0.0	0.0%	0.0	0.0%
Other	944	568.2	1.4%	548.6	1.3%

---

Top 20 Tasks by Self-Time

Self-time is the exclusive duration: time spent in the task itself, not in sub-tasks.
This is the most accurate indicator of where CPU cycles are actually spent.

Self (ms)	Inclusive (ms)	Count	Avg Self (us)	P95 Self (ms)	Max Self (ms)	% Work	Task Name	Top Caller
13,917.9	18,822.2	130,533	106.6	0.1	38.6	34.2%	module	write all entrypoints to disk (1%)
4,379.5	4,416.0	26,375	166.0	0.2	257.2	10.8%	analyze ecmascript module	module (67%)
3,128.0	3,345.6	40,833	76.6	0.0	25.8	7.7%	internal resolving	resolving (31%)
3,001.0	6,548.3	38,075	78.8	0.0	26.8	7.4%	process module	module (14%)
2,608.4	3,793.7	33,186	78.6	0.0	19.7	6.4%	resolving	module (15%)
2,326.3	2,495.0	6,392	363.9	1.1	45.4	5.7%	parse ecmascript	process module (35%)
1,629.7	2,614.5	16,258	100.2	0.4	60.2	4.0%	code generation	chunking (1%)
1,199.2	1,308.8	11,120	107.8	0.3	74.7	2.9%	chunking	write all entrypoints to disk (0%)
887.6	1,290.7	4,325	205.2	0.1	141.9	2.2%	write all entrypoints to disk	None (0%)
827.6	827.6	9,404	88.0	0.3	9.5	2.0%	precompute code generation	code generation (60%)
698.8	698.8	5,460	128.0	0.4	88.6	1.7%	compute async module info	chunking (0%)
515.6	516.0	2,921	176.5	0.2	36.5	1.3%	compute async chunks	webpack loader (0%)
499.9	513.3	857	583.3	3.6	27.1	1.2%	webpack loader	parse css (17%)
344.0	647.7	662	519.6	2.2	36.1	0.8%	parse css	module (7%)
322.5	322.5	2,084	154.7	0.4	16.4	0.8%	generate source map	code generation (96%)
175.3	175.3	2,340	74.9	0.0	18.1	0.4%	read file	parse ecmascript (91%)
59.0	59.0	694	85.0	0.0	15.8	0.1%	compute binding usage info	write all entrypoints to disk (0%)
43.1	59.4	513	83.9	0.1	6.1	0.1%	async reference	write all entrypoints to disk (1%)
23.4	53.0	62	377.5	0.6	17.3	0.1%	make production chunks	chunking (6%)
22.4	22.4	36	621.0	3.3	7.1	0.1%	write file	apply effects (100%)

---

Critical Path Analysis

The longest sequential dependency chains that determine wall-clock time.
Focus on reducing the depth of these chains to improve parallelism.

Rank	Self-Time (ms)	Depth	Path
1	257.3	2	process module → analyze ecmascript module
2	80.0	2	module → analyze ecmascript module
3	76.6	2	code generation → generate source map
4	50.3	2	process module → analyze ecmascript module
5	45.4	2	process module → parse ecmascript

---

Batching Candidates

High-volume tasks dominated by a single parent. If the parent can batch them,
it drastically reduces scheduler overhead.

Task Name	Count	Top Caller (Attribution)	Avg Self	P95 Self	Total Self
No obvious batching candidates found	-	-	-	-	-

---

Duration Distribution

Range	Count	Percentage
<10us	1,232,722	78.7%
10us-100us	307,385	19.6%
100us-1ms	18,215	1.2%
1ms-10ms	6,940	0.4%
10ms-100ms	703	0.0%
>100ms	6	0.0%

---

Action Items

1. [P0] Focus on tasks with the highest Self-Time — these are where CPU cycles are actually spent.
2. [P0] Use Batching Candidates to identify callers that should use try_join or reduce #[turbo_tasks::function] granularity.
3. [P1] Check Build Phase Timeline for phases with disproportionate wall range vs. self-time (= serialization).
4. [P1] Inspect P95 Self (ms) for heavy monolith tasks. Focus on long-tail outliers, not averages.
5. [P1] Review Critical Paths — reducing the longest chain depth directly improves wall-clock time.
6. [P2] If Thread Utilization < 60%, investigate scheduling gaps (lock contention or deep dependency chains).

Report generated by Utoopack Performance Analysis Agent