afpd: harden build_fce_packet() remaining accounting

[rdmark]

build_fce_packet() had imperfect remaining accounting: after writing 16-bit length fields for username/path/source path, it advances p and datalen but does not subtract those length-field bytes from remaining

also fixes a straggler FCE packet version 1 marshalling bug, a leftover from the fix for SourceForge Bug No.603

[augmentcode]

🤖 Augment PR Summary

Summary: Fixes incorrect buffer-remaining accounting in build_fce_packet() so the 16-bit length fields are included when decrementing remaining space.

Changes:

• Subtracts sizeof(uint16) + payload length for username/path/source-path fields to prevent under-accounting
• Passes obj->fce_version (instead of the compile-time constant) to the notify script to keep version marshalling consistent for v1/v2

_{🤖 Was this summary useful? React with 👍 or 👎}

[augmentcode]

Review completed. No suggestions at this time.

Comment augment review to trigger a new review at any time.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[github-actions]

📊 Performance Dashboard

Commit: 1cb628d39622c15dc7d23b54632ba414a7d18528

🔥 Spectest (AFP 3.4) - FlameGraph

Netatalk Code-time: 3.6% · Runtime: 68s · Stacks: 1895

🔥 Click the preview to open the interactive flamegraph (zoom + search).

🔝 Top 10 leaf functions

Function	Samples
do_syscall_64	395810065
x64_sys_call	248070375
_raw_spin_unlock_irqrestore	217199395
finish_task_switch.isra.0	192943625
__cp_end	98125615
srso_alias_safe_ret	79382520
afpd	68357170
find_get_block_common	38588725
__pi_memset	30870980
kmem_cache_alloc_noprof	27563375

📈 Speedtest (AFP 3.4) - PerfGraph

🔝 Peak throughput per operation

Operation	Peak mean (MB/s)
Read	6327
Write	1568
Copy	2393
ServerCopy	3254

⏱️ Lantest (AFP 3.4) - LatencyGraph

Avg total runtime: 4342 ms · Avg time per AFP op: 0.084 ms

🐢 All operations (avg runtime, in test order)

Test	Avg (ms)
Writing one large file	37
Reading one large file	16
Creating 2000 files	701
Writing 1024 bytes to 2000 files	372
Lock then unlock 2000 open forks	153
Stat, open, read 512 bytes, close 2000 files	773
Enumerate dir with 2000 files	5
Deleting 2000 files	629
Byte-range lock/unlock 2000 ranges in one fork	157
Create directory tree with 1000 dirs	396
Directory cache hits (20 dirs x 100 files)	92
Mixed cache operations (create/stat/enum/delete) on 1000 files	646
Deep path traversal (20 levels x 100 walks)	99
Cache validation (2000 files x 3 lookups)	266