[FEA] Strongly filtered CAGRA

CAGRA has been observed to yield low recall when filtering is enabled, especially when the ratio of filtered-out values is high. This can be related in part to #208 and #472 , but there also may be fundamental reasons for the lower recall.

This feature request tracks the progress and suggestions to enable high-recall strongly filtered CAGRA.

As an experiment, I suggest to try the following tweaks, enabled by a boolean search parameter:

• Disable the maximum search iterations limit to allow longer search
• Replace the hashmap with a dataset-long bitset. It's used to track the visited nodes. By replacing a small hashmap with the bitset we will eliminate hash collisions (thus, false-positives) and prevent CAGRA from early-stopping.

Start with cpp/include/cuvs/neighbors/cagra.hpp and cpp/src/neighbors/detail/cagra/hashmap.hpp to understand the existing search-iteration limit, search parameters, and visited-node tracking. Compare the proposed parameter-controlled alternatives and establish how high recall under strong filtering would be measured; done requires an agreed design and validation results.