docs: fix typos in Kandinsky 5.0 video docs

[iridescentWen]

What does this PR do?

Fixes spelling mistakes in the Kandinsky 5.0 Video docs. Prose and comments only — no code or behavior changes.

qualty → quality	line 16
infered → inferred	lines 57, 152, 170
bakend → backend	lines 68, 129, 181
wothout → without	line 152

The bakend hits are all in the trailing # <--- Set attention bakend to Flex comments; the actual set_attention_backend(...) calls were already spelled correctly.

Docs-only: 1 file, +7/-7.

Verification

No Python source touched (git diff --name-only shows only the .md), so there is nothing for make quality to check. Instead:

• codespell on the file is clean after the change (7 hits before, 0 after).
• Line count and triple-backtick fence count are identical to main (310 lines, 13 fences — odd because of the inline ```guidance_scale=1.0``` on the distilled-model line), so the markdown structure is untouched.
• All 5 ```python blocks still parse with ast.parse.

Self-review notes

Ran the self-review skill against .ai/review-rules.md (plus AGENTS.md) on git diff upstream/main...HEAD.

Blocking issues: none.

Non-blocking issues: none.

Dead code: N/A — docs-only diff, no code paths added or removed.

Notes / deliberate calls:

1. Kept this PR to spelling only. While checking the file I noticed the runnable examples have separate problems that are not addressed here — happy to open an issue or a follow-up PR if you want them fixed:

   • The Image-to-Video example loads the ...I2V-Pro... checkpoint with Kandinsky5T2VPipeline, calls load_image(...) without importing it, and never passes the resulting image to pipe(...) even though image is the first (required) parameter of Kandinsky5I2VPipeline.__call__.
   • The Pro T2V and I2V examples assign pipe = ... and then call pipeline.transformer... / pipeline.enable_model_cpu_offload(), which is a NameError on copy-paste.

   I left all of it alone to keep this a single-purpose spelling diff rather than mixing a correctness fix into it.

2. Scoped to this one file. codespell reports ~125 hits across the English docs and src/, e.g. sequencially ×15 in modular_pipelines/qwenimage/denoise.py and exlucde in components_manager.py. Those want a separate systematic sweep, not a bundle with this one. Let me know if a repo-wide pass would be welcome.

3. Left the two prose pipeline.enable_model_cpu_offload() mentions as-is. They read as "call this method" rather than as copy-paste code, and renaming them is out of scope for a spelling fix.

Verdict: READY

Before submitting

• Did you use an AI agent (Claude Code, Codex, Cursor, etc.) to help with this PR? If so:
  • Did you read the Coding with AI agents guide?
  • Did you run the self-review skill on the diff?
  • Did you share the final self-review notes in the PR description or a comment?
• Did you read the contributor guideline?
• Did you make sure to update the documentation with your changes?
• Did you write any new necessary tests? — N/A, docs-only.
• Are you the author (or part of the team) of the model/pipeline? — No.

AI disclosure: this PR was prepared with Claude Code. I verified every change myself and checked each spelling against the surrounding text.

Who can review?

@stevhliu — docs-only spelling fix.

[stevhliu]

Thanks for fixing the typos, feel free to include others here as well!

[github-actions]

Hi @iridescentWen, thanks for the PR! It does not appear to link an issue it fixes. If this PR addresses an existing issue, please add a closing keyword (e.g. Fixes #1234) to the PR description so the issue is linked. See the contribution guide for more details. If this PR intentionally does not fix a tracked issue, a maintainer can add the no-issue-needed label to silence this reminder.