feat: Depend on dplyr instead of reexporting all generics

[krlmlr]

This ensures that requesting help for mutate() jumps to the dplyr help page, and not to the disambiguation page.

[github-actions]

This is how benchmark results would change (along with a 95% confidence interval in relative change) if 0df1d4b is merged into main:

• ✔️001_tpch_01: 24.3ms -> 24.1ms [-5.12%, +3.37%]
• ❗🐌001_tpch_02: 131ms -> 164ms [+23.23%, +26.05%]
• ✔️001_tpch_03: 68.3ms -> 68.1ms [-2.31%, +1.5%]
• ✔️001_tpch_04: 23ms -> 22.6ms [-3.78%, +0.36%]
• ❗🐌001_tpch_05: 125ms -> 159ms [+25.34%, +28.14%]
• ❗🐌001_tpch_06: 13.9ms -> 20.1ms [+41.76%, +46.99%]
• ✔️001_tpch_07: 143ms -> 142ms [-1.84%, +0.48%]
• ❗🐌001_tpch_08: 92.1ms -> 160ms [+72.45%, +74.67%]
• ✔️001_tpch_09: 134ms -> 135ms [-0.71%, +1.83%]
• 🚀001_tpch_10: 113ms -> 74.8ms [-34.67%, -32.28%]
• 🚀001_tpch_11: 65ms -> 33.1ms [-50.36%, -47.82%]
• 🚀001_tpch_12: 59.8ms -> 27.7ms [-54.88%, -52.4%]
• ❗🐌001_tpch_13: 19.7ms -> 25.6ms [+27.59%, +31.65%]
• ❗🐌001_tpch_14: 20.6ms -> 26.4ms [+25.8%, +30.41%]
• 🚀001_tpch_15: 66.4ms -> 32.3ms [-52.82%, -49.93%]
• ✔️001_tpch_16: 68.6ms -> 68.1ms [-2.86%, +1.33%]
• 🚀001_tpch_17: 58.6ms -> 26.8ms [-55.55%, -53.01%]
• ✔️001_tpch_18: 23.1ms -> 23.3ms [-1.64%, +3.04%]
• 🚀001_tpch_19: 134ms -> 68.6ms [-49.65%, -47.54%]
• ✔️001_tpch_20: 80.7ms -> 79.8ms [-3.02%, +0.84%]
• 🚀001_tpch_21: 146ms -> 106ms [-28.11%, -26.12%]
• 🚀001_tpch_22: 131ms -> 95.8ms [-27.63%, -25.75%]
• ✔️010_tpch_01: 79.6ms -> 81.6ms [-0.98%, +6.01%]
• ✔️010_tpch_02: 71.9ms -> 72.7ms [-0.99%, +3.35%]
• ✔️010_tpch_03: 61.3ms -> 61.8ms [-1.01%, +2.57%]
• ✔️010_tpch_04: 44.2ms -> 44.1ms [-4.4%, +3.95%]
• ✔️010_tpch_05: 93.2ms -> 92.5ms [-2.73%, +1.2%]
• ✔️010_tpch_06: 32.6ms -> 33ms [-5.7%, +8.41%]
• ✔️010_tpch_07: 111ms -> 109ms [-4.73%, +2.18%]
• ✔️010_tpch_08: 128ms -> 130ms [-1.45%, +3.7%]
• ✔️010_tpch_09: 116ms -> 118ms [-1.16%, +5.01%]
• ✔️010_tpch_10: 76.6ms -> 77.7ms [-0.06%, +2.92%]
• ✔️010_tpch_11: 39.2ms -> 39.3ms [-3.45%, +3.94%]
• ✔️010_tpch_12: 58.4ms -> 59.8ms [-5.49%, +10.23%]
• ✔️010_tpch_13: 52.9ms -> 53.2ms [-1.55%, +2.63%]
• ✔️010_tpch_14: 37.7ms -> 38.1ms [-0.46%, +2.56%]
• ✔️010_tpch_15: 55.7ms -> 55.4ms [-7.84%, +7.03%]
• ✔️010_tpch_16: 46.2ms -> 46.3ms [-2.57%, +3.16%]
• ✔️010_tpch_17: 54.1ms -> 56ms [-1.03%, +7.99%]
• ✔️010_tpch_18: 53.4ms -> 55.3ms [-4.24%, +11.27%]
• ✔️010_tpch_19: 117ms -> 117ms [-0.4%, +1.23%]
• ✔️010_tpch_20: 70ms -> 68.9ms [-4.87%, +1.98%]
• ✔️010_tpch_21: 242ms -> 240ms [-3.39%, +1.88%]
• ✔️010_tpch_22: 78.4ms -> 77.6ms [-2.59%, +0.33%]
• ✔️100_tpch_01: 314ms -> 306ms [-19.17%, +13.95%]
• ✔️100_tpch_02: 127ms -> 127ms [-5.23%, +5.02%]
• ✔️100_tpch_03: 184ms -> 184ms [-4.67%, +4.55%]
• ✔️100_tpch_04: 154ms -> 160ms [-5.44%, +14.08%]
• ✔️100_tpch_05: 287ms -> 265ms [-17.79%, +2.48%]
• ✔️100_tpch_06: 101ms -> 104ms [-7.05%, +14.14%]
• ✔️100_tpch_07: 241ms -> 243ms [-7.81%, +9.38%]
• ✔️100_tpch_08: 254ms -> 255ms [-12.31%, +12.53%]
• ✔️100_tpch_09: 365ms -> 336ms [-27.74%, +11.96%]
• ✔️100_tpch_10: 214ms -> 217ms [-7.79%, +10.77%]
• ✔️100_tpch_11: 82.1ms -> 84.3ms [-2.9%, +8.42%]
• ✔️100_tpch_12: 195ms -> 200ms [-12.57%, +16.94%]
• ✔️100_tpch_13: 342ms -> 326ms [-18.42%, +9.31%]
• ✔️100_tpch_14: 124ms -> 131ms [-4.44%, +14.66%]
• ✔️100_tpch_15: 218ms -> 220ms [-11.69%, +12.86%]
• ✔️100_tpch_16: 126ms -> 132ms [-11.72%, +20.73%]
• ✔️100_tpch_17: 184ms -> 188ms [-3.55%, +7.27%]
• ✔️100_tpch_18: 204ms -> 199ms [-12.49%, +7.89%]
• ✔️100_tpch_19: 283ms -> 291ms [-7.24%, +12.88%]
• ✔️100_tpch_20: 182ms -> 180ms [-4.91%, +3.09%]
• ✔️100_tpch_21: 1.31s -> 1.33s [-0.77%, +3.12%]
• ✔️100_tpch_22: 175ms -> 175ms [-11.59%, +11.5%]

Further explanation regarding interpretation and methodology can be found in the documentation.