Significant TokenMap optimisation by removing clones and returning refs instead.

[JoshuaBatty]

Description

This PR refactors the functions implemented on TokenMap to remove clones where possible, and instead return the RefMulti type from DashMap directly. As can be seen below, we are getting significant performance wins and most LSP capabilities are now measured in µs. 🚀

related to #5445

Metric	Before	After	Change (%)
tokens_for_file	9.8117 ms	253.845 µs	-97.411%
idents_at_position	3.6331 ms	47.870 µs	-98.692%
tokens_at_position	6.9532 ms	1.1478 ms	-83.474%
token_at_position	5.9395 ms	208.23 µs	-96.483%
parent_decl_at_position	6.9780 ms	1.1633 ms	-83.428%
semantic_tokens	18.2205 ms	587.64 µs	-96.767%
document_symbol	6.6865 ms	584.33 µs	-91.186%
completion	14.0025 ms	1.9808 ms	-85.855%
hover	6.2409 ms	317.90 µs	-94.902%
highlight	14.8390 ms	2.7572 ms	-81.417%
goto_definition	5.9403 ms	225.65 µs	-96.204%
inlay_hints	5.9934 ms	318.41 µs	-94.677%
prepare_rename	5.8056 ms	207.62 µs	-96.411%
rename	12.5200 ms	2.8592 ms	-77.168%
code_action	11.0798 ms	1.9683 ms	-82.233%

Checklist

• I have linked to any relevant issues.
• I have commented my code, particularly in hard-to-understand areas.
• I have updated the documentation where relevant (API docs, the reference, and the Sway book).
• I have added tests that prove my fix is effective or that my feature works.
• I have added (or requested a maintainer to add) the necessary Breaking* or New Feature labels where relevant.
• I have done my best to ensure that my PR adheres to the Fuel Labs Code Review Standards.
• I have requested a review from the relevant team or maintainers.

[sdankel]

These performance gains are fantastic! Nice work 🚀

[tritao]

Nice work, amazing performance gains 🚀