elf: fix R_386_PC8 relocation patching on x86-32

[TharunBabu-05]

Your checklist for this pull request

• I've read the guidelines for contributing to this repository.
• I made sure to follow the project's coding style.
• I've documented every RZ_API function and struct this PR changes. (No RZ_API changes — the edited patch_reloc_x86_32 is static.)
• I've added tests that prove my changes are effective.
• I've updated the Rizin book with the relevant information (not needed).
• I've used AI tools to generate fully or partially these code changes and I'm sure the changes are not copyrighted by somebody else.

Detailed description

In patch_reloc_x86_32() (librz/bin/format/elf/elf_relocs_patching.c), the
R_386_8 / R_386_PC8 case subtracted the place address P only when
rel_type == R_386_PC16 — a copy-paste slip from the R_386_16 / R_386_PC16
case directly above it. So an R_386_PC8 relocation was patched to S + A
instead of the PC-relative S + A - P, writing the wrong byte. R_386_8
(which must stay S + A) was unaffected. The x86-64 patcher already handles
R_X86_64_PC8 correctly.

This is part of the x86 relocation gaps tracked in #5259.

AI disclosure: Claude Code (Anthropic Claude) was used to find the bug and
draft the fix and test; I reviewed and verified everything. The commit carries
the Co-authored-by agent: trailer per AGENTS.md.

Test plan

Added a case to test/db/formats/elf/elf-rel-x86 using a small object with an
R_386_PC8 relocation (and an R_386_8 control), added in
rizinorg/rizin-testbins#6841 as elf/x86_relocs_pc8.o. It reads the
patched bytes:

• R_386_PC8 site @ 0x08000035 → 0x02 (= 0x08000037 - 0x08000035)
• R_386_8 site @ 0x08000036 → 0x37 (unchanged)

Verified on CI: with the fix rz-test db/formats/elf/elf-rel-x86 → 2 OK;
with the fix reverted → 1 XX (expected 02, actual 37), confirming the
test catches the regression.

Closing issues

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

[TharunBabu-05]

Your checklist for this pull request

• I've read the guidelines for contributing to this repository.
• I made sure to follow the project's coding style.
• I've documented every RZ_API function and struct this PR changes. (No RZ_API changes — the edited patch_reloc_x86_32 is static.)
• I've added tests that prove my changes are effective.
• I've updated the Rizin book with the relevant information (not needed).
• I've used AI tools to generate fully or partially these code changes and I'm sure the changes are not copyrighted by somebody else.

Detailed description

In patch_reloc_x86_32() (librz/bin/format/elf/elf_relocs_patching.c), the
R_386_8 / R_386_PC8 case subtracted the place address P only when
rel_type == R_386_PC16 — a copy-paste slip from the R_386_16 / R_386_PC16
case directly above it. So an R_386_PC8 relocation was patched to S + A
instead of the PC-relative S + A - P, writing the wrong byte. R_386_8
(which must stay S + A) was unaffected. The x86-64 patcher already handles
R_X86_64_PC8 correctly.

This is part of the x86 relocation gaps tracked in #5259.

AI disclosure: Claude Code (Anthropic Claude) was used to find the bug and
draft the fix and test; I reviewed and verified everything. The commit carries
the Co-authored-by agent: trailer per AGENTS.md.

Test plan

Added a case to test/db/formats/elf/elf-rel-x86 using a small object with an
R_386_PC8 relocation (and an R_386_8 control), added in
rizinorg/rizin-testbins#314 as elf/x86_relocs_pc8.o. It reads the
patched bytes:

• R_386_PC8 site @ 0x08000035 → 0x02 (= 0x08000037 - 0x08000035)
• R_386_8 site @ 0x08000036 → 0x37 (unchanged)

Verified on CI: with the fix rz-test db/formats/elf/elf-rel-x86 → 2 OK;
with the fix reverted → 1 XX (expected 02, actual 37), confirming the
test catches the regression.

Closing issues

[Rot127]

Please rewrite the PR description by hand and give a reference to the ABI specs where that is documented.