Limit AMF string and array parsing

[wwbmmm]

What problem does this PR solve?

Issue Number: null

Problem Summary:

AMF parsing accepted declared string lengths and array counts directly from the input stream. Malformed RTMP/AMF payloads could make the parser reserve excessive memory or spend a long time walking impossible element counts before discovering that the stream was incomplete.

What is changed and the side effects?

Changed:

• Add configurable AMF string and array limits.
• Read AMF string bodies incrementally instead of resizing the destination string to the declared length up front.
• Reject ECMA array and strict array counts above the configured limit before parsing their elements.
• Add RTMP AMF unit tests for truncated long strings and oversized array counts.

Side effects:

• Performance effects: AMF string parsing now copies through a small stack buffer, avoiding large zero-fill allocations for malformed payloads.

• Breaking backward compatibility: AMF payloads whose string length or array count exceed the new defaults will be rejected. The defaults are 64 MiB per string and 1,048,576 array elements.

---

Check List:

• Please make sure your changes are compilable.
• When providing us with a new feature, it is best to add related tests.
• Please follow Contributor Covenant Code of Conduct.

[chenBright]

LGTM