Use values for topology target selectors

[SandroMaglione]

Summary

• make definition-time topology instructions immutable values while keeping state, choice, and runtime branch evidence builders callable
• adopt target-first machine initial entry with once-only selector capture and lazy resolver execution
• remove Machine.targetless in favor of (to) => to.none, migrate the repository, and preserve exact transition/lifecycle inference and compiled metadata
• reject active or choice child keys that collide with the reserved initial and schema-backed local with selector properties

Changeset

• Added or updated for a library or package-metadata change
• Not required because this PR does not change src/ or package.json

Validation

• pnpm check
• Relevant example checks, when examples changed
• Automated type-performance measurement passed or was not required
• Automated runtime- and memory-performance measurement passed or was not required

Affected example checks passed for platformer, playground, and pokemon. Local type-performance budgets and the runtime/memory smoke suite passed; CI base-versus-PR reports remain authoritative.

[github-actions]

Type performance

Measured with TypeScript 6.0.3 and skipLibCheck=true.

Scenario	Base	PR	Difference
Effect only	55	55	0 (0.0%)
Import effect-machine	55	55	0 (0.0%)
Machine.states (3 states)	3,039	3,039	0 (0.0%)
Machine.make (3 states, 2 events)	10,439	10,761	+322 (+3.1%)
machine.handle (3 states, 2 transitions)	32,036	29,177	-2,859 (-8.9%)
fluent transition (10 named branches)	119,749	109,470	-10,279 (-8.6%)
Machine.invoke (state-dependent Effect)	93,494	92,024	-1,470 (-1.6%)
machine.handle (depth 24)	205,744	198,478	-7,266 (-3.5%)
machine.handle (wide depth 16)	237,098	229,624	-7,474 (-3.2%)
machine.handle (parallel/history/choice)	141,115	133,024	-8,091 (-5.7%)
machine definition (3 independent implementations)	140,991	132,018	-8,973 (-6.4%)
machine exact input/output/error/services	126,285	117,186	-9,099 (-7.2%)
execution adapter readiness	135,186	126,635	-8,551 (-6.3%)

Marginal instantiations are measured against the matching setup without that API call:

Scenario	Base	PR	Difference
Import effect-machine	0	0	0
Machine.states (3 states)	2,984	2,984	0 (0.0%)
Machine.make (3 states, 2 events)	7,392	7,714	+322 (+4.4%)
machine.handle (3 states, 2 transitions)	21,597	18,416	-3,181 (-14.7%)
fluent transition (10 named branches)	109,820	99,219	-10,601 (-9.7%)
Machine.invoke (state-dependent Effect)	84,648	82,757	-1,891 (-2.2%)
machine.handle (depth 24)	186,054	177,780	-8,274 (-4.4%)
machine.handle (wide depth 16)	218,407	210,133	-8,274 (-3.8%)
machine.handle (parallel/history/choice)	121,439	112,757	-8,682 (-7.1%)
machine definition (3 independent implementations)	125,445	116,045	-9,400 (-7.5%)
machine exact input/output/error/services	112,194	102,673	-9,521 (-8.5%)
execution adapter readiness	103,709	95,462	-8,247 (-8.0%)

Check times (informational)

Scenario	Base	PR
Effect only	0.02s	0.02s
Import effect-machine	0.02s	0.02s
Machine.states (3 states)	0.07s	0.07s
Machine.make (3 states, 2 events)	0.12s	0.12s
machine.handle (3 states, 2 transitions)	0.20s	0.20s
fluent transition (10 named branches)	0.41s	0.40s
Machine.invoke (state-dependent Effect)	0.36s	0.35s
machine.handle (depth 24)	0.61s	0.59s
machine.handle (wide depth 16)	0.61s	0.60s
machine.handle (parallel/history/choice)	0.48s	0.46s
machine definition (3 independent implementations)	0.49s	0.47s
machine exact input/output/error/services	0.45s	0.44s
execution adapter readiness	0.46s	0.45s

Type instantiations are the comparison metric. Check time varies with runner load and is informational only.

[github-actions]

Runtime performance

Median of 5 independent benchmark processes on AMD EPYC 7763 64-Core Processor with Node v24.19.0.

Pull request baseline

Scenario	Effect Machine
Plan counter transitions	114,007 transitions/s
Drain burst with terminal fence	345,430 increments/s
Drain burst with a change observer	324,903 increments/s
Lookup and send to one child	286,387 increments/s
Start and stop a machine	140,766 machines/s
Start and stop a parent with one child	27,452 families/s
Plan transitions through a compound state	98,063 transitions/s
Plan transitions through parallel regions	80,661 transitions/s
Drain burst through a compound state	297,448 events/s
Drain burst through two parallel regions	300,114 events/s
Drain a compound-state burst with a change observer	278,298 events/s

Process runtime reference points

Scenario	Effect Machine
Start and stop a raw generic process	15,156 processes/s
Start and stop a raw compiled process	61,308 processes/s

Memory profile	Effect Machine
Idle machine	1.8 KiB
Raw generic managed process	13.9 KiB
Raw compiled process	3.1 KiB
Two independent idle machines	3.4 KiB
Idle parent with one child	5.8 KiB
Parent with observed child registry	10.0 KiB
Parent with observed invoked child snapshots	6.3 KiB

Effect Machine change from base

Metric	Base	Base variability	PR	PR variability	Difference
Plan counter transitions	107,731 transitions/s	0.7% MAD	114,007 transitions/s	1.6% MAD	+5.8%
Drain burst with terminal fence	332,803 increments/s	0.9% MAD	345,430 increments/s	0.4% MAD	+3.8%
Drain burst with a change observer	313,145 increments/s	1.2% MAD	324,903 increments/s	0.3% MAD	+3.8%
Lookup and send to one child	274,758 increments/s	1.5% MAD	286,387 increments/s	0.1% MAD	+4.2%
Start and stop a machine	138,447 machines/s	0.6% MAD	140,766 machines/s	0.3% MAD	+1.7%
Start and stop a parent with one child	27,803 families/s	4.1% MAD	27,452 families/s	1.8% MAD	-1.3%
Plan transitions through a compound state	94,311 transitions/s	1.0% MAD	98,063 transitions/s	2.8% MAD	+4.0%
Plan transitions through parallel regions	76,112 transitions/s	0.8% MAD	80,661 transitions/s	0.4% MAD	+6.0%
Drain burst through a compound state	289,514 events/s	1.9% MAD	297,448 events/s	0.8% MAD	+2.7%
Drain burst through two parallel regions	290,644 events/s	0.9% MAD	300,114 events/s	0.5% MAD	+3.3%
Drain a compound-state burst with a change observer	272,812 events/s	1.4% MAD	278,298 events/s	0.6% MAD	+2.0%
Idle machine heap per unit	1.8 KiB	0.0% MAD	1.8 KiB	0.2% MAD	-0.0%
Raw generic managed process heap per unit	13.9 KiB	0.0% MAD	13.9 KiB	0.0% MAD	-0.0%
Raw compiled process heap per unit	3.1 KiB	0.1% MAD	3.1 KiB	0.2% MAD	+0.2%
Two independent idle machines heap per unit	3.4 KiB	0.0% MAD	3.4 KiB	0.1% MAD	-0.0%
Idle parent with one child heap per unit	5.8 KiB	0.0% MAD	5.8 KiB	0.0% MAD	-0.0%
Parent with observed child registry heap per unit	10.0 KiB	0.0% MAD	10.0 KiB	0.0% MAD	-0.0%
Parent with observed invoked child snapshots heap per unit	6.3 KiB	0.0% MAD	6.3 KiB	0.0% MAD	+0.1%

Process runtime reference change from base

Metric	Base	Base variability	PR	PR variability	Difference
Start and stop a raw generic process	15,114 processes/s	3.9% MAD	15,156 processes/s	2.3% MAD	+0.3%
Start and stop a raw compiled process	60,533 processes/s	1.1% MAD	61,308 processes/s	0.9% MAD	+1.3%

Regression guard

No large, noise-adjusted throughput or heap regressions detected.

Versions and interpretation

• Effect Machine: 0.16.0

Higher throughput is better; lower heap is better. Variability is the median absolute deviation across independent processes, relative to their median. Small differences on shared GitHub-hosted hardware remain informational; the required guard rejects only large changes beyond the measured noise allowance.