tpl/collections: Use MapRange/SetIterKey/SetIterValue for Where, Sort and Merge

tpl/collections: Use MapRange/SetIterKey/SetIterValue for Where, Sort and Merge

Some relevant benchmarks:

Where with maps:

```
cpu: Apple M1 Pro
            │ master.bench │         fix-mapkeys.bench          │
            │    sec/op    │   sec/op     vs base               │
WhereMap-10    79.26µ ± 1%   26.58µ ± 1%  -66.46% (p=0.002 n=6)

            │ master.bench │         fix-mapkeys.bench         │
            │     B/op     │    B/op     vs base               │
WhereMap-10   56685.0 ± 0%   111.0 ± 1%  -99.80% (p=0.002 n=6)

            │ master.bench  │         fix-mapkeys.bench         │
            │   allocs/op   │ allocs/op   vs base               │
WhereMap-10   2003.000 ± 0%   4.000 ± 0%  -99.80% (p=0.002 n=6)
```

Merge:

```
         │ master.bench │         fix-mapkeys.bench          │
         │    sec/op    │   sec/op     vs base               │
Merge-10    3.285µ ± 0%   2.268µ ± 1%  -30.96% (p=0.002 n=6)

         │ master.bench │          fix-mapkeys.bench          │
         │     B/op     │     B/op      vs base               │
Merge-10   3.079Ki ± 0%   1.891Ki ± 0%  -38.58% (p=0.002 n=6)

         │ master.bench │         fix-mapkeys.bench         │
         │  allocs/op   │ allocs/op   vs base               │
Merge-10     64.00 ± 0%   26.00 ± 0%  -59.38% (p=0.002 n=6)
```
Sort:

```
cpu: Apple M1 Pro
           │ master.bench │         fix-mapkeys.bench         │
           │    sec/op    │   sec/op     vs base              │
SortMap-10   1008.0n ± 1%   915.5n ± 0%  -9.18% (p=0.002 n=6)

           │ master.bench │         fix-mapkeys.bench         │
           │     B/op     │    B/op     vs base               │
SortMap-10     640.0 ± 0%   512.0 ± 0%  -20.00% (p=0.002 n=6)

           │ master.bench │        fix-mapkeys.bench         │
           │  allocs/op   │ allocs/op   vs base              │
SortMap-10     16.00 ± 0%   15.00 ± 0%  -6.25% (p=0.002 n=6)
```