Optimize `path`, `join`, and `removeNullOrUndef` functions for performance and clarity, introduce regex pre-compilation, and improve path parameter validation. #82

Performance Gate Passed

⚡ 19 improvements
✅ 139 untouched
⏩ 22 skipped¹

	Benchmark	`BASE`	`HEAD`	Change
⚡	`Join with separator at end of first part`	17 µs	15.4 µs	+10.62%
⚡	`Join URL with trailing slash`	16.8 µs	14.4 µs	+16.23%
⚡	`Complex URL with many params (original test)`	182.2 µs	141.3 µs	+28.97%
⚡	`Deep path with 5 parameters`	118.4 µs	100.8 µs	+17.49%
⚡	`Large query string (15 params)`	151.5 µs	104.8 µs	+44.55%
⚡	`Large query string (15 params) with some path params`	166.6 µs	120.3 µs	+38.47%
⚡	`Array in query params`	84.7 µs	67.5 µs	+25.46%
⚡	`Base template only (no path)`	98.2 µs	80.5 µs	+22.09%
⚡	`Boolean and number params`	113.8 µs	94.4 µs	+20.5%
⚡	`Multiple query params (5 params)`	111.2 µs	86.4 µs	+28.69%
⚡	`Path params with special characters`	113.1 µs	96.4 µs	+17.34%
⚡	`Two path params + two query params`	112.6 µs	100.6 µs	+11.92%
⚡	`REST API endpoint with pagination`	85.7 µs	74.8 µs	+14.64%
⚡	`Resource with ID and nested resource`	101.8 µs	84.8 µs	+20.05%
⚡	`Search endpoint with filters`	96.1 µs	72.6 µs	+32.37%
⚡	`URL with port number`	99.8 µs	82.3 µs	+21.18%
⚡	`Base URL with trailing slash + path`	31 µs	25.4 µs	+21.89%
⚡	`Concatenate base URL and path only`	76.6 µs	67.1 µs	+14.02%
⚡	`Single query parameter`	92.9 µs	72.5 µs	+28.13%

22 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩